Add name and description to v1/models list (#179)

* Add support for name and description in v1/models list
* add configuration example for name and description

.github/ISSUE_TEMPLATE/bug-report.md

@@ -0,0 +1,37 @@
+---
+name: Bug Report
+about: Something is not working as expected...
+title: ''
+labels: bug
+assignees: ''
+
+---
+
+**Describe the bug**
+A clear and concise description of what the bug is.
+
+**Expected behaviour**
+A clear and concise description of what you expected to happen.
+
+**Operating system and version**
+
+- OS: (linux, osx, windows, freebsd, etc)
+- GPUs: (list architecture)
+
+**My Configuration**
+
+```yaml
+# copy / paste your configuration here
+```
+
+**Proxy Logs**
+
+```
+# copy / paste from /logs
+```
+
+**Upstream Logs**
+
+```
+# copy/paste from /logs
+```

.github/workflows/go-ci-windows.yml

@@ -0,0 +1,50 @@
+name: Windows CI
+
+on:
+  push:
+    branches: [ "main" ]
+
+  pull_request:
+    branches: [ "main" ]
+
+  # Allows manual triggering of the workflow
+  workflow_dispatch:
+
+jobs:
+
+  run-tests:
+    runs-on: windows-latest
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Go
+      uses: actions/setup-go@v4
+      with:
+        go-version: '1.23'
+
+    # cache simple-responder to save the build time
+    - name: Restore Simple Responder
+      id: restore-simple-responder
+      uses: actions/cache/restore@v4
+      with:
+        path: ./build
+        key: ${{ runner.os }}-simple-responder-${{ hashFiles('misc/simple-responder/simple-responder.go') }}
+
+    # necessary for testing proxy/Process swapping
+    - name: Create simple-responder
+      if: steps.restore-simple-responder.outputs.cache-hit != 'true'
+      shell: bash
+      run: make simple-responder-windows
+
+    - name: Save Simple Responder
+      # nothing new to save ... skip this step
+      if: steps.restore-simple-responder.outputs.cache-hit != 'true'
+      id: save-simple-responder
+      uses: actions/cache/save@v4
+      with:
+        path: ./build
+        key: ${{ runner.os }}-simple-responder-${{ hashFiles('misc/simple-responder/simple-responder.go') }}
+
+    - name: Test all
+      shell: bash
+      run: make test-all

.github/workflows/go-ci.yml

@@ -1,6 +1,4 @@
-# This workflow will build a golang project
-
-name: CI
+name: Linux CI
 
 on:
   push:
@@ -24,9 +22,26 @@ jobs:
       with:
         go-version: '1.23'
 
+    # cache simple-responder to save the build time
+    - name: Restore Simple Responder
+      id: restore-simple-responder
+      uses: actions/cache/restore@v4
+      with:
+        path: ./build
+        key: ${{ runner.os }}-simple-responder-${{ hashFiles('misc/simple-responder/simple-responder.go') }}
+
     # necessary for testing proxy/Process swapping
     - name: Create simple-responder
       run: make simple-responder
 
+    - name: Save Simple Responder
+      # nothing new to save ... skip this step
+      if: steps.restore-simple-responder.outputs.cache-hit != 'true'
+      id: save-simple-responder
+      uses: actions/cache/save@v4
+      with:
+        path: ./build
+        key: ${{ runner.os }}-simple-responder-${{ hashFiles('misc/simple-responder/simple-responder.go') }}
+
     - name: Test all
       run: make test-all

.github/workflows/release.yml

@@ -23,6 +23,19 @@ jobs:
       -
         name: Set up Go
         uses: actions/setup-go@v5
+
+      -
+        name: Set up Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '23'  # or your preferred version
+      -
+        name: Install dependencies and build UI
+        run: |
+          cd ui
+          npm ci
+          npm run build
+
       -
         name: Run GoReleaser
         uses: goreleaser/goreleaser-action@v6

.goreleaser.yaml

@@ -17,14 +17,16 @@ builds:
       - goos: windows
         goarch: arm64
 
-# use zip format for windows
 archives:
   - id: default
-    format: tar.gz
+    formats:
+      - tar.gz
     name_template: "{{ .ProjectName }}_{{ .Version }}_{{ .Os }}_{{ .Arch }}"
     builds_info:
       group: root
       owner: root
     format_overrides:
+      # use zip format for windows
       - goos: windows
-        format: zip
+        formats:
+          - zip

Makefile

@@ -19,24 +19,35 @@ all: mac linux simple-responder
 clean:
 	rm -rf $(BUILD_DIR)
 
-test:
-	go test -short -v ./proxy
+proxy/ui_dist/placeholder.txt:
+	mkdir -p proxy/ui_dist
+	touch $@
 
-test-all:
-	go test -v ./proxy
+test: proxy/ui_dist/placeholder.txt
+	go test -short -v -count=1 ./proxy
+
+test-all: proxy/ui_dist/placeholder.txt
+	go test -v -count=1 ./proxy
+
+ui/node_modules:
+	cd ui && npm install
+
+# build react UI
+ui: ui/node_modules
+	cd ui && npm run build
 
 # Build OSX binary
-mac:
+mac: ui
 	@echo "Building Mac binary..."
 	GOOS=darwin GOARCH=arm64 go build -ldflags="-X main.commit=${GIT_HASH} -X main.version=local_${GIT_HASH} -X main.date=${BUILD_DATE}" -o $(BUILD_DIR)/$(APP_NAME)-darwin-arm64
 
 # Build Linux binary
-linux:
+linux: ui
 	@echo "Building Linux binary..."
 	GOOS=linux GOARCH=amd64 go build -ldflags="-X main.commit=${GIT_HASH} -X main.version=local_${GIT_HASH} -X main.date=${BUILD_DATE}" -o $(BUILD_DIR)/$(APP_NAME)-linux-amd64
 
 # Build Windows binary
-windows:
+windows: ui
 	@echo "Building Windows binary..."
 	GOOS=windows GOARCH=amd64 go build -ldflags="-X main.commit=${GIT_HASH} -X main.version=local_${GIT_HASH} -X main.date=${BUILD_DATE}" -o $(BUILD_DIR)/$(APP_NAME)-windows-amd64.exe
 
@@ -46,6 +57,10 @@ simple-responder:
 	GOOS=darwin GOARCH=arm64 go build -o $(BUILD_DIR)/simple-responder_darwin_arm64 misc/simple-responder/simple-responder.go
 	GOOS=linux GOARCH=amd64 go build -o $(BUILD_DIR)/simple-responder_linux_amd64 misc/simple-responder/simple-responder.go
 
+simple-responder-windows:
+	@echo "Building simple responder for windows"
+	GOOS=windows GOARCH=amd64 go build -o $(BUILD_DIR)/simple-responder.exe misc/simple-responder/simple-responder.go
+
 # Ensure build directory exists
 $(BUILD_DIR):
 	mkdir -p $(BUILD_DIR)
@@ -65,4 +80,4 @@ release:
 	git tag "$$new_tag";
 
 # Phony targets
-.PHONY: all clean mac linux windows simple-responder
+.PHONY: all clean ui mac linux windows simple-responder

README.md

@@ -7,7 +7,7 @@
 
 llama-swap is a light weight, transparent proxy server that provides automatic model swapping to llama.cpp's server.
 
-Written in golang, it is very easy to install (single binary with no dependancies) and configure (single yaml file). To get started, download a pre-built binary or use the provided docker images.
+Written in golang, it is very easy to install (single binary with no dependencies) and configure (single yaml file). To get started, download a pre-built binary or use the provided docker images.
 
 ## Features:
 
@@ -22,6 +22,7 @@ Written in golang, it is very easy to install (single binary with no dependancie
   - `v1/audio/speech` ([#36](https://github.com/mostlygeek/llama-swap/issues/36))
   - `v1/audio/transcriptions` ([docs](https://github.com/mostlygeek/llama-swap/issues/41#issuecomment-2722637867))
 - ✅ llama-swap custom API endpoints
+  - `/ui` - web UI
   - `/log` - remote log monitoring
   - `/upstream/:model_id` - direct access to upstream HTTP server ([demo](https://github.com/mostlygeek/llama-swap/pull/31))
   - `/unload` - manually unload running models ([#58](https://github.com/mostlygeek/llama-swap/issues/58))
@@ -40,166 +41,47 @@ In the most basic configuration llama-swap handles one model at a time. For more
 
 ## config.yaml
 
-llama-swap's configuration is purposefully simple.
+llama-swap is managed entirely through a yaml configuration file. 
+
+It can be very minimal to start: 
 
 ```yaml
 models:
   "qwen2.5":
-    proxy: "http://127.0.0.1:9999"
-    cmd: >
-      /app/llama-server
+    cmd: |
+      /path/to/llama-server
       -hf bartowski/Qwen2.5-0.5B-Instruct-GGUF:Q4_K_M
-      --port 9999
-
-  "smollm2":
-    proxy: "http://127.0.0.1:9999"
-    cmd: >
-      /app/llama-server
-      -hf bartowski/SmolLM2-135M-Instruct-GGUF:Q4_K_M
-      --port 9999
+      --port ${PORT}
 ```
 
-<details>
-<summary>But also very powerful ...</summary>
+However, there are many more capabilities that llama-swap supports: 
 
-```yaml
-# Seconds to wait for llama.cpp to load and be ready to serve requests
-# Default (and minimum) is 15 seconds
-healthCheckTimeout: 60
+- `groups` to run multiple models at once
+- `ttl` to automatically unload models
+- `macros` for reusable snippets
+- `aliases` to use familiar model names (e.g., "gpt-4o-mini")
+- `env` to pass custom environment variables to inference servers
+- `cmdStop` for to gracefully stop Docker/Podman containers
+- `useModelName` to override model names sent to upstream servers
+- `healthCheckTimeout` to control model startup wait times
+- `${PORT}` automatic port variables for dynamic port assignment
 
-# Valid log levels: debug, info (default), warn, error
-logLevel: info
+See the [configuration documentation](https://github.com/mostlygeek/llama-swap/wiki/Configuration) in the wiki all options and examples.
 
-# Automatic Port Values
-# use ${PORT} in model.cmd and model.proxy to use an automatic port number
-# when you use ${PORT} you can omit a custom model.proxy value, as it will
-# default to http://localhost:${PORT}
+## Web UI
 
-# override the default port (5800) for automatic port values
-startPort: 10001
+llama-swap ships with a web based interface to make it easier to monitor logs and check the status of models. 
 
-# define valid model values and the upstream server start
-models:
-  "llama":
-    # multiline for readability
-    cmd: >
-      llama-server --port 8999
-      --model path/to/Qwen2.5-1.5B-Instruct-Q4_K_M.gguf
-
-    # environment variables to pass to the command
-    env:
-      - "CUDA_VISIBLE_DEVICES=0"
-
-    # where to reach the server started by cmd, make sure the ports match
-    # can be omitted if you use an automatic ${PORT} in cmd
-    proxy: http://127.0.0.1:8999
-
-    # aliases names to use this model for
-    aliases:
-      - "gpt-4o-mini"
-      - "gpt-3.5-turbo"
-
-    # check this path for an HTTP 200 OK before serving requests
-    # default: /health to match llama.cpp
-    # use "none" to skip endpoint checking, but may cause HTTP errors
-    # until the model is ready
-    checkEndpoint: /custom-endpoint
-
-    # automatically unload the model after this many seconds
-    # ttl values must be a value greater than 0
-    # default: 0 = never unload model
-    ttl: 60
-
-    # `useModelName` overrides the model name in the request
-    # and sends a specific name to the upstream server
-    useModelName: "qwen:qwq"
-
-  # unlisted models do not show up in /v1/models or /upstream lists
-  # but they can still be requested as normal
-  "qwen-unlisted":
-    unlisted: true
-    cmd: llama-server --port ${PORT} -m Llama-3.2-1B-Instruct-Q4_K_M.gguf -ngl 0
-
-  # Docker Support (v26.1.4+ required!)
-  "docker-llama":
-    proxy: "http://127.0.0.1:${PORT}"
-    cmd: >
-      docker run --name dockertest
-      --init --rm -p ${PORT}:8080 -v /mnt/nvme/models:/models
-      ghcr.io/ggerganov/llama.cpp:server
-      --model '/models/Qwen2.5-Coder-0.5B-Instruct-Q4_K_M.gguf'
-
-# Groups provide advanced controls over model swapping behaviour. Using groups
-# some models can be kept loaded indefinitely, while others are swapped out.
-#
-# Tips:
-#
-#  - models must be defined above in the Models section
-#  - a model can only be a member of one group
-#  - group behaviour is controlled via the `swap`, `exclusive` and `persistent` fields
-#  - see issue #109 for details
-#
-# NOTE: the example below uses model names that are not defined above for demonstration purposes
-groups:
-  # group1 is the default behaviour of llama-swap where only one model is allowed
-  # to run a time across the whole llama-swap instance
-  "group1":
-    # swap controls the model swapping behaviour in within the group
-    # - true : only one model is allowed to run at a time
-    # - false: all models can run together, no swapping
-    swap: true
-
-    # exclusive controls how the group affects other groups
-    # - true: causes all other groups to unload their models when this group runs a model
-    # - false: does not affect other groups
-    exclusive: true
-
-    # members references the models defined above
-    members:
-      - "llama"
-      - "qwen-unlisted"
-
-  # models in this group are never unloaded
-  "group2":
-    swap: false
-    exclusive: false
-    members:
-      - "docker-llama"
-      # (not defined above, here for example)
-      - "modelA"
-      - "modelB"
-
-  "forever":
-    # setting persistent to true causes the group to never be affected by the swapping behaviour of
-    # other groups. It is a shortcut to keeping some models always loaded.
-    persistent: true
-
-    # set swap/exclusive to false to prevent swapping inside the group and effect on other groups
-    swap: false
-    exclusive: false
-    members:
-      - "forever-modelA"
-      - "forever-modelB"
-      - "forever-modelc"
-```
-
-### Use Case Examples
-
-- [config.example.yaml](config.example.yaml) includes example for supporting `v1/embeddings` and `v1/rerank` endpoints
-- [Speculative Decoding](examples/speculative-decoding/README.md) - using a small draft model can increase inference speeds from 20% to 40%. This example includes a configurations Qwen2.5-Coder-32B (2.5x increase) and Llama-3.1-70B (1.4x increase) in the best cases.
-- [Optimizing Code Generation](examples/benchmark-snakegame/README.md) - find the optimal settings for your machine. This example demonstrates defining multiple configurations and testing which one is fastest.
-- [Restart on Config Change](examples/restart-on-config-change/README.md) - automatically restart llama-swap when trying out different configurations.
-</details>
+<img width="1758" alt="image" src="https://github.com/user-attachments/assets/31ae5bcd-5efd-46b0-b64b-6db9e60196d3" />
 
 ## Docker Install ([download images](https://github.com/mostlygeek/llama-swap/pkgs/container/llama-swap))
 
 Docker is the quickest way to try out llama-swap:
 
 ```shell
-# use CPU inference
+# use CPU inference comes with the example config above
 $ docker run -it --rm -p 9292:8080 ghcr.io/mostlygeek/llama-swap:cpu
 
-
 # qwen2.5 0.5B
 $ curl -s http://localhost:9292/v1/chat/completions \
     -H "Content-Type: application/json" \
@@ -207,7 +89,6 @@ $ curl -s http://localhost:9292/v1/chat/completions \
     -d '{"model":"qwen2.5","messages": [{"role": "user","content": "tell me a joke"}]}' | \
     jq -r '.choices[0].message.content'
 
-
 # SmolLM2 135M
 $ curl -s http://localhost:9292/v1/chat/completions \
     -H "Content-Type: application/json" \
@@ -217,7 +98,7 @@ $ curl -s http://localhost:9292/v1/chat/completions \
 ```
 
 <details>
-<summary>Docker images are nightly ...</summary>
+<summary>Docker images are built nightly for cuda, intel, vulcan, etc ...</summary>
 
 They include:
 
@@ -242,29 +123,29 @@ $ docker run -it --rm --runtime nvidia -p 9292:8080 \
 
 ## Bare metal Install ([download](https://github.com/mostlygeek/llama-swap/releases))
 
-Pre-built binaries are available for Linux, FreeBSD and Darwin (OSX). These are automatically published and are likely a few hours ahead of the docker releases. The baremetal install works with any OpenAI compatible server, not just llama-server.
+Pre-built binaries are available for Linux, Mac, Windows and FreeBSD. These are automatically published and are likely a few hours ahead of the docker releases. The baremetal install works with any OpenAI compatible server, not just llama-server.
 
-1. Create a configuration file, see [config.example.yaml](config.example.yaml)
 1. Download a [release](https://github.com/mostlygeek/llama-swap/releases) appropriate for your OS and architecture.
-1. Run the binary with `llama-swap --config path/to/config.yaml`.
-  Available flags:
-    - `--config`: Path to the configuration file (default: `config.yaml`).
-    - `--listen`: Address and port to listen on (default: `:8080`).
-    - `--version`: Show version information and exit.
-    - `--watch-config`: Automatically reload the configuration file when it changes. This will wait for in-flight requests to complete then stop all running models (default: `false`).
+1. Create a configuration file, see the [configuration documentation](https://github.com/mostlygeek/llama-swap/wiki/Configuration).
+1. Run the binary with `llama-swap --config path/to/config.yaml --listen localhost:8080`.
+   Available flags:
+   - `--config`: Path to the configuration file (default: `config.yaml`).
+   - `--listen`: Address and port to listen on (default: `:8080`).
+   - `--version`: Show version information and exit.
+   - `--watch-config`: Automatically reload the configuration file when it changes. This will wait for in-flight requests to complete then stop all running models (default: `false`).
 
 ### Building from source
 
-1. Install golang for your system
+1. Build requires golang and nodejs for the user interface.
 1. `git clone git@github.com:mostlygeek/llama-swap.git`
 1. `make clean all`
 1. Binaries will be in `build/` subdirectory
 
 ## Monitoring Logs
 
-Open the `http://<host>/logs` with your browser to get a web interface with streaming logs.
+Open the `http://<host>:<port>/` with your browser to get a web interface with streaming logs.
 
-Of course, CLI access is also supported:
+CLI access is also supported:
 
 ```shell
 # sends up to the last 10KB of logs
@@ -292,32 +173,6 @@ Any OpenAI compatible server would work. llama-swap was originally designed for
 
 For Python based inference servers like vllm or tabbyAPI it is recommended to run them via podman or docker. This provides clean environment isolation as well as responding correctly to `SIGTERM` signals to shutdown.
 
-## Systemd Unit Files
-
-Use this unit file to start llama-swap on boot. This is only tested on Ubuntu.
-
-`/etc/systemd/system/llama-swap.service`
-
-```
-[Unit]
-Description=llama-swap
-After=network.target
-
-[Service]
-User=nobody
-
-# set this to match your environment
-ExecStart=/path/to/llama-swap --config /path/to/llama-swap.config.yml
-
-Restart=on-failure
-RestartSec=3
-StartLimitBurst=3
-StartLimitInterval=30
-
-[Install]
-WantedBy=multi-user.target
-```
-
 ## Star History
 
 [![Star History Chart](https://api.star-history.com/svg?repos=mostlygeek/llama-swap&type=Date)](https://www.star-history.com/#mostlygeek/llama-swap&Date)

config.example.yaml

@@ -1,88 +1,203 @@
-# Seconds to wait for llama.cpp to be available to serve requests
-# Default (and minimum): 15 seconds
-healthCheckTimeout: 90
+# llama-swap YAML configuration example
+# -------------------------------------
+#
+# - Below are all the available configuration options for llama-swap.
+# - Settings with a default value, or noted as optional can be omitted.
+# - Settings that are marked required must be in your configuration file
 
-# valid log levels: debug, info (default), warn, error
-logLevel: debug
+# healthCheckTimeout: number of seconds to wait for a model to be ready to serve requests
+# - optional, default: 120
+# - minimum value is 15 seconds, anything less will be set to this value
+healthCheckTimeout: 500
 
-# creating a coding profile with models for code generation and general questions
-groups:
-  coding:
-    swap: false
-    members:
-      - "qwen"
-      - "llama"
+# logLevel: sets the logging value
+# - optional, default: info
+# - Valid log levels: debug, info, warn, error
+logLevel: info
 
+# startPort: sets the starting port number for the automatic ${PORT} macro.
+# - optional, default: 5800
+# - the ${PORT} macro can be used in model.cmd and model.proxy settings
+# - it is automatically incremented for every model that uses it
+startPort: 10001
+
+# macros: sets a dictionary of string:string pairs
+# - optional, default: empty dictionary
+# - these are reusable snippets
+# - used in a model's cmd, cmdStop, proxy and checkEndpoint
+# - useful for reducing common configuration settings
+macros:
+  "latest-llama": >
+    /path/to/llama-server/llama-server-ec9e0301
+    --port ${PORT}
+
+# models: a dictionary of model configurations
+# - required
+# - each key is the model's ID, used in API requests
+# - model settings have default values that are used if they are not defined here
+# - below are examples of the various settings a model can have:
+# - available model settings: env, cmd, cmdStop, proxy, aliases, checkEndpoint, ttl, unlisted
 models:
+
+  # keys are the model names used in API requests
   "llama":
-    cmd: >
-      models/llama-server-osx
-      --port ${PORT}
-      -m models/Llama-3.2-1B-Instruct-Q4_0.gguf
+    # cmd: the command to run to start the inference server.
+    # - required
+    # - it is just a string, similar to what you would run on the CLI
+    # - using `|` allows for comments in the command, these will be parsed out
+    # - macros can be used within cmd
+    cmd: |
+      # ${latest-llama} is a macro that is defined above
+      ${latest-llama}
+      --model path/to/llama-8B-Q4_K_M.gguf
 
-    # list of model name aliases this llama.cpp instance can serve
+    # name: a display name for the model
+    # - optional, default: empty string
+    # - if set, it will be used in the v1/models API response
+    # - if not set, it will be omitted in the JSON model record
+    name: "llama 3.1 8B"
+
+    # description: a description for the model
+    # - optional, default: empty string
+    # - if set, it will be used in the v1/models API response
+    # - if not set, it will be omitted in the JSON model record
+    description: "A small but capable model used for quick testing"
+
+    # env: define an array of environment variables to inject into cmd's environment
+    # - optional, default: empty array
+    # - each value is a single string
+    # - in the format: ENV_NAME=value
+    env:
+      - "CUDA_VISIBLE_DEVICES=0,1,2"
+
+    # proxy: the URL where llama-swap routes API requests
+    # - optional, default: http://localhost:${PORT}
+    # - if you used ${PORT} in cmd this can be omitted
+    # - if you use a custom port in cmd this *must* be set
+    proxy: http://127.0.0.1:8999
+
+    # aliases: alternative model names that this model configuration is used for
+    # - optional, default: empty array
+    # - aliases must be unique globally
+    # - useful for impersonating a specific model
     aliases:
-    - gpt-4o-mini
+      - "gpt-4o-mini"
+      - "gpt-3.5-turbo"
 
-    # check this path for a HTTP 200 response for the server to be ready
-    checkEndpoint: /health
+    # checkEndpoint: URL path to check if the server is ready
+    # - optional, default: /health
+    # - use "none" to skip endpoint ready checking
+    # - endpoint is expected to return an HTTP 200 response
+    # - all requests wait until the endpoint is ready (or fails)
+    checkEndpoint: /custom-endpoint
 
-    # unload model after 5 seconds
-    ttl: 5
+    # ttl: automatically unload the model after this many seconds
+    # - optional, default: 0
+    # - ttl values must be a value greater than 0
+    # - a value of 0 disables automatic unloading of the model
+    ttl: 60
 
-  "qwen":
-    cmd: models/llama-server-osx --port ${PORT} -m models/qwen2.5-0.5b-instruct-q8_0.gguf
-    aliases:
-      - gpt-3.5-turbo
+    # useModelName: overrides the model name that is sent to upstream server
+    # - optional, default: ""
+    # - useful when the upstream server expects a specific model name or format
+    useModelName: "qwen:qwq"
 
-  # Embedding example with Nomic
-  # https://huggingface.co/nomic-ai/nomic-embed-text-v1.5-GGUF
-  "nomic":
-    cmd: >
-      models/llama-server-osx --port ${PORT}
-      -m models/nomic-embed-text-v1.5.Q8_0.gguf
-      --ctx-size 8192
-      --batch-size 8192
-      --rope-scaling yarn
-      --rope-freq-scale 0.75
-      -ngl 99
-      --embeddings
+    # filters: a dictionary of filter settings
+    # - optional, default: empty dictionary
+    filters:
+      # strip_params: a comma separated list of parameters to remove from the request
+      # - optional, default: ""
+      # - useful for preventing overriding of default server params by requests
+      # - `model` parameter is never removed
+      # - can be any JSON key in the request body
+      # - recommended to stick to sampling parameters
+      strip_params: "temperature, top_p, top_k"
 
-  # Reranking example with bge-reranker
-  # https://huggingface.co/gpustack/bge-reranker-v2-m3-GGUF
-  "bge-reranker":
-    cmd: >
-      models/llama-server-osx --port ${PORT}
-      -m models/bge-reranker-v2-m3-Q4_K_M.gguf
-      --ctx-size 8192
-      --reranking
+  # Unlisted model example:
+  "qwen-unlisted":
+    # unlisted: true or false
+    # - optional, default: false
+    # - unlisted models do not show up in /v1/models or /upstream lists
+    # - can be requested as normal through all apis
+    unlisted: true
+    cmd: llama-server --port ${PORT} -m Llama-3.2-1B-Instruct-Q4_K_M.gguf -ngl 0
 
-  # Docker Support (v26.1.4+ required!)
-  "dockertest":
-    cmd: >
+  # Docker example:
+  # container run times like Docker and Podman can also be used with a
+  # a combination of cmd and cmdStop.
+  "docker-llama":
+    proxy: "http://127.0.0.1:${PORT}"
+    cmd: |
       docker run --name dockertest
       --init --rm -p ${PORT}:8080 -v /mnt/nvme/models:/models
-      ghcr.io/ggerganov/llama.cpp:server
+      ghcr.io/ggml-org/llama.cpp:server
       --model '/models/Qwen2.5-Coder-0.5B-Instruct-Q4_K_M.gguf'
 
-  "simple":
-    # example of setting environment variables
-    env:
-      - CUDA_VISIBLE_DEVICES=0,1
-      - env1=hello
-    cmd: build/simple-responder --port ${PORT}
-    unlisted: true
+    # cmdStop: command to run to stop the model gracefully
+    # - optional, default: ""
+    # - useful for stopping commands managed by another system
+    # - on POSIX systems: a SIGTERM is sent for graceful shutdown
+    # - on Windows, taskkill is used
+    # - processes are given 5 seconds to shutdown until they are forcefully killed
+    # - the upstream's process id is available in the ${PID} macro
+    cmdStop: docker stop dockertest
 
-    # use "none" to skip check. Caution this may cause some requests to fail
-    # until the upstream server is ready for traffic
-    checkEndpoint: none
+# groups: a dictionary of group settings
+# - optional, default: empty dictionary
+# - provide advanced controls over model swapping behaviour.
+# - Using groups some models can be kept loaded indefinitely, while others are swapped out.
+# - model ids must be defined in the Models section
+# - a model can only be a member of one group
+# - group behaviour is controlled via the `swap`, `exclusive` and `persistent` fields
+# - see issue #109 for details
+#
+# NOTE: the example below uses model names that are not defined above for demonstration purposes
+groups:
+  # group1 is same as the default behaviour of llama-swap where only one model is allowed
+  # to run a time across the whole llama-swap instance
+  "group1":
+    # swap: controls the model swapping behaviour in within the group
+    # - optional, default: true
+    # - true : only one model is allowed to run at a time
+    # - false: all models can run together, no swapping
+    swap: true
 
-  # don't use these, just for testing if things are broken
-  "broken":
-    cmd: models/llama-server-osx --port 8999 -m models/doesnotexist.gguf
-    proxy: http://127.0.0.1:8999
-    unlisted: true
-  "broken_timeout":
-    cmd: models/llama-server-osx --port 8999 -m models/qwen2.5-0.5b-instruct-q8_0.gguf
-    proxy: http://127.0.0.1:9000
-    unlisted: true
+    # exclusive: controls how the group affects other groups
+    # - optional, default: true
+    # - true: causes all other groups to unload when this group runs a model
+    # - false: does not affect other groups
+    exclusive: true
+
+    # members references the models defined above
+    # required
+    members:
+      - "llama"
+      - "qwen-unlisted"
+
+  # Example:
+  # - in this group all the models can run at the same time
+  # - when a different group loads all running models in this group are unloaded
+  "group2":
+    swap: false
+    exclusive: false
+    members:
+      - "docker-llama"
+      - "modelA"
+      - "modelB"
+
+  # Example:
+  # - a persistent group, prevents other groups from unloading it
+  "forever":
+    # persistent: prevents over groups from unloading the models in this group
+    # - optional, default: false
+    # - does not affect individual model behaviour
+    persistent: true
+
+    # set swap/exclusive to false to prevent swapping inside the group
+    # and the unloading of other groups
+    swap: false
+    exclusive: false
+    members:
+      - "forever-modelA"
+      - "forever-modelB"
+      - "forever-modelc"

go.mod

@@ -5,7 +5,6 @@ go 1.23.0
 require (
 	github.com/fsnotify/fsnotify v1.9.0
 	github.com/gin-gonic/gin v1.10.0
-	github.com/google/shlex v0.0.0-20191202100458-e7afc7fbc510
 	github.com/stretchr/testify v1.9.0
 	github.com/tidwall/gjson v1.18.0
 	github.com/tidwall/sjson v1.2.5
@@ -13,6 +12,7 @@ require (
 )
 
 require (
+	github.com/billziss-gh/golib v0.2.0 // indirect
 	github.com/bytedance/sonic v1.11.6 // indirect
 	github.com/bytedance/sonic/loader v0.1.1 // indirect
 	github.com/cloudwego/base64x v0.1.4 // indirect

go.sum

@@ -1,3 +1,5 @@
+github.com/billziss-gh/golib v0.2.0 h1:NyvcAQdfvM8xokKkKotiligKjKXzuQD4PPykg1nKc/8=
+github.com/billziss-gh/golib v0.2.0/go.mod h1:mZpUYANXZkDKSnyYbX9gfnyxwe0ddRhUtfXcsD5r8dw=
 github.com/bytedance/sonic v1.11.6 h1:oUp34TzMlL+OY1OUWxHqsdkgC/Zfc85zGqw9siXjrc0=
 github.com/bytedance/sonic v1.11.6/go.mod h1:LysEHSvpvDySVdC2f87zGWf6CIKJcAvqab1ZaiQtds4=
 github.com/bytedance/sonic/loader v0.1.1 h1:c+e5Pt1k/cy5wMveRDyk2X4B9hF4g7an8N3zCYjJFNM=

llama-swap.go

@@ -144,8 +144,8 @@ func watchConfigFileWithReload(configPath string, reloadChan chan<- *proxy.Proxy
 			if !ok {
 				return
 			}
-			// We only care about writes to the specific config file
-			if event.Name == configPath && event.Has(fsnotify.Write) {
+			// We only care about writes/creates to the specific config file
+			if event.Name == configPath && (event.Has(fsnotify.Write) || event.Has(fsnotify.Create) || event.Has(fsnotify.Remove)) {
 				// Reset or start the debounce timer
 				if debounceTimer != nil {
 					debounceTimer.Stop()
@@ -176,6 +176,13 @@ func watchConfigFileWithReload(configPath string, reloadChan chan<- *proxy.Proxy
 					newPM := proxy.New(newConfig)
 					reloadChan <- newPM
 					log.Println("Config reloaded successfully")
+					if (event.Has(fsnotify.Remove)) {
+						// re-add watcher
+						err = watcher.Add(configPath)
+						if err != nil {
+							log.Printf("Could not re-add watcher for %s: %s", configPath, err)
+						}
+					}
 				})
 			}
 		case err, ok := <-watcher.Errors:

misc/process-cmd-test/main.go

@@ -0,0 +1,91 @@
+package main
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"os"
+	"os/exec"
+	"os/signal"
+	"syscall"
+	"time"
+)
+
+/*
+**
+Test how exec.Cmd.CommandContext behaves under certain conditions:*
+
+  - process is killed externally, what happens with cmd.Wait() *
+    ✔︎ it returns. catches crashes.*
+
+  - process ignores SIGTERM*
+    ✔︎ `kill()` is called after cmd.WaitDelay*
+
+  - this process exits, what happens with children (kill -9 <this process' pid>)*
+    x they stick around. have to be manually killed.*
+
+  - .WithTimeout()'s cancel is called *
+    ✔︎ process is killed after it ignores sigterm, cmd.Wait() catches it.*
+
+  - parent receives SIGINT/SIGTERM, what happens
+    ✔︎ waits for child process to exit, then exits gracefully.
+*/
+func main() {
+
+	// swap between these to use kill -9 <pid> on the cli to sim external crash
+	ctx, cancel := context.WithCancel(context.Background())
+	//ctx, cancel := context.WithTimeout(context.Background(), 1000*time.Millisecond)
+	defer cancel()
+
+	//cmd := exec.CommandContext(ctx, "sleep", "1")
+	cmd := exec.CommandContext(ctx,
+		"../../build/simple-responder_darwin_arm64",
+		//"-ignore-sig-term", /* so it doesn't exit on receiving SIGTERM, test cmd.WaitTimeout */
+	)
+	cmd.Stdin = os.Stdin
+	cmd.Stdout = os.Stdout
+	cmd.Stderr = os.Stderr
+
+	// set a wait delay before signing sig kill
+	cmd.WaitDelay = 500 * time.Millisecond
+	cmd.Cancel = func() error {
+		fmt.Println("✔︎ Cancel() called, sending SIGTERM")
+		cmd.Process.Signal(syscall.SIGTERM)
+
+		//return nil
+
+		// this error is returned by cmd.Wait(), and can be used to
+		// single an error when the process couldn't be normally terminated
+		// but since a SIGTERM is sent, it's probably ok to return a nil
+		// as WaitDelay timing out will override the any error set here.
+		//
+		// test by enabling/disabling -ignore-sig-term on the process
+		// with -ignore-sig-term enabled, cmd.Wait() will have "signal: killed"
+		// without it, it will show the "new error from cancel"
+		return errors.New("error from cmd.Cancel()") // sets error returned by cmd.Wait()
+	}
+
+	if err := cmd.Start(); err != nil {
+		fmt.Println("Error starting process:", err)
+		return
+	}
+
+	// catch signals. Calls cancel() which will cause cmd.Wait() to return and
+	// this program to eventually exit gracefully.
+	sigChan := make(chan os.Signal, 1)
+	signal.Notify(sigChan, syscall.SIGINT, syscall.SIGTERM)
+	go func() {
+		signal := <-sigChan
+		fmt.Printf("✔︎ Received signal: %d, Killing process... with cancel before exiting\n", signal)
+		cancel()
+	}()
+
+	fmt.Printf("✔︎ Parent Pid: %d, Process Pid: %d\n", os.Getpid(), cmd.Process.Pid)
+	fmt.Println("✔︎ Process started, cmd.Wait() ... ")
+	if err := cmd.Wait(); err != nil {
+		fmt.Println("✔︎ cmd.Wait returned, Error:", err)
+	} else {
+		fmt.Println("✔︎ cmd.Wait returned, Process exited on its own")
+	}
+	fmt.Println("✔︎ Child process exited, Done.")
+}

misc/simple-responder/simple-responder.go

@@ -26,6 +26,8 @@ func main() {
 
 	silent := flag.Bool("silent", false, "disable all logging")
 
+	ignoreSigTerm := flag.Bool("ignore-sig-term", false, "ignore SIGTERM signal")
+
 	flag.Parse() // Parse the command-line flags
 
 	// Create a new Gin router
@@ -40,9 +42,12 @@ func main() {
 			time.Sleep(wait)
 		}
 
+		bodyBytes, _ := io.ReadAll(c.Request.Body)
+
 		c.JSON(http.StatusOK, gin.H{
 			"responseMessage":  *responseMessage,
 			"h_content_length": c.Request.Header.Get("Content-Length"),
+			"request_body":     string(bodyBytes),
 		})
 	})
 
@@ -190,6 +195,10 @@ func main() {
 		log.SetOutput(io.Discard)
 	}
 
+	if !*silent {
+		fmt.Printf("My PID: %d\n", os.Getpid())
+	}
+
 	go func() {
 		log.Printf("simple-responder listening on %s\n", address)
 		// service connections
@@ -200,11 +209,36 @@ func main() {
 
 	// Wait for interrupt signal to gracefully shutdown the server with
 	// a timeout of 5 seconds.
-	quit := make(chan os.Signal, 1)
+	sigChan := make(chan os.Signal, 1)
 	// kill (no param) default send syscall.SIGTERM
 	// kill -2 is syscall.SIGINT
 	// kill -9 is syscall.SIGKILL but can't be catch, so don't need add it
-	signal.Notify(quit, syscall.SIGINT, syscall.SIGTERM)
-	<-quit
+	signal.Notify(sigChan, syscall.SIGINT, syscall.SIGTERM)
+
+	countSigInt := 0
+
+runloop:
+	for {
+		signal := <-sigChan
+		switch signal {
+		case syscall.SIGINT:
+			countSigInt++
+			if countSigInt > 1 {
+				break runloop
+			} else {
+				log.Println("Received SIGINT, send another SIGINT to shutdown")
+			}
+		case syscall.SIGTERM:
+			if *ignoreSigTerm {
+				log.Println("Ignoring SIGTERM")
+			} else {
+				log.Println("Received SIGTERM, shutting down")
+				break runloop
+			}
+		default:
+			break runloop
+		}
+	}
+
 	log.Println("simple-responder shutting down")
 }

proxy/.gitignore

@@ -0,0 +1 @@
+ui_dist/*

proxy/config.go

@@ -4,11 +4,14 @@ import (
 	"fmt"
 	"io"
 	"os"
+	"regexp"
+	"runtime"
+	"slices"
 	"sort"
 	"strconv"
 	"strings"
 
-	"github.com/google/shlex"
+	"github.com/billziss-gh/golib/shlex"
 	"gopkg.in/yaml.v3"
 )
 
@@ -16,6 +19,7 @@ const DEFAULT_GROUP_ID = "(default)"
 
 type ModelConfig struct {
 	Cmd           string   `yaml:"cmd"`
+	CmdStop       string   `yaml:"cmdStop"`
 	Proxy         string   `yaml:"proxy"`
 	Aliases       []string `yaml:"aliases"`
 	Env           []string `yaml:"env"`
@@ -24,14 +28,91 @@ type ModelConfig struct {
 	Unlisted      bool     `yaml:"unlisted"`
 	UseModelName  string   `yaml:"useModelName"`
 
+	// #179 for /v1/models
+	Name        string `yaml:"name"`
+	Description string `yaml:"description"`
+
 	// Limit concurrency of HTTP requests to process
 	ConcurrencyLimit int `yaml:"concurrencyLimit"`
+
+	// Model filters see issue #174
+	Filters ModelFilters `yaml:"filters"`
+}
+
+func (m *ModelConfig) UnmarshalYAML(unmarshal func(interface{}) error) error {
+	type rawModelConfig ModelConfig
+	defaults := rawModelConfig{
+		Cmd:              "",
+		CmdStop:          "",
+		Proxy:            "http://localhost:${PORT}",
+		Aliases:          []string{},
+		Env:              []string{},
+		CheckEndpoint:    "/health",
+		UnloadAfter:      0,
+		Unlisted:         false,
+		UseModelName:     "",
+		ConcurrencyLimit: 0,
+		Name:             "",
+		Description:      "",
+	}
+
+	// the default cmdStop to taskkill /f /t /pid ${PID}
+	if runtime.GOOS == "windows" {
+		defaults.CmdStop = "taskkill /f /t /pid ${PID}"
+	}
+
+	if err := unmarshal(&defaults); err != nil {
+		return err
+	}
+
+	*m = ModelConfig(defaults)
+	return nil
 }
 
 func (m *ModelConfig) SanitizedCommand() ([]string, error) {
 	return SanitizeCommand(m.Cmd)
 }
 
+// ModelFilters see issue #174
+type ModelFilters struct {
+	StripParams string `yaml:"strip_params"`
+}
+
+func (m *ModelFilters) UnmarshalYAML(unmarshal func(interface{}) error) error {
+	type rawModelFilters ModelFilters
+	defaults := rawModelFilters{
+		StripParams: "",
+	}
+
+	if err := unmarshal(&defaults); err != nil {
+		return err
+	}
+
+	*m = ModelFilters(defaults)
+	return nil
+}
+
+func (f ModelFilters) SanitizedStripParams() ([]string, error) {
+	if f.StripParams == "" {
+		return nil, nil
+	}
+
+	params := strings.Split(f.StripParams, ",")
+	cleaned := make([]string, 0, len(params))
+
+	for _, param := range params {
+		trimmed := strings.TrimSpace(param)
+		if trimmed == "model" || trimmed == "" {
+			continue
+		}
+		cleaned = append(cleaned, trimmed)
+	}
+
+	// sort cleaned
+	slices.Sort(cleaned)
+	return cleaned, nil
+}
+
 type GroupConfig struct {
 	Swap       bool     `yaml:"swap"`
 	Exclusive  bool     `yaml:"exclusive"`
@@ -65,6 +146,9 @@ type Config struct {
 	Profiles           map[string][]string    `yaml:"profiles"`
 	Groups             map[string]GroupConfig `yaml:"groups"` /* key is group ID */
 
+	// for key/value replacements in model's cmd, cmdStop, proxy, checkEndPoint
+	Macros map[string]string `yaml:"macros"`
+
 	// map aliases to actual model IDs
 	aliases map[string]string
 
@@ -105,21 +189,23 @@ func LoadConfigFromReader(r io.Reader) (Config, error) {
 		return Config{}, err
 	}
 
-	var config Config
+	// default configuration values
+	config := Config{
+		HealthCheckTimeout: 120,
+		StartPort:          5800,
+		LogLevel:           "info",
+	}
 	err = yaml.Unmarshal(data, &config)
 	if err != nil {
 		return Config{}, err
 	}
 
 	if config.HealthCheckTimeout < 15 {
+		// set a minimum of 15 seconds
 		config.HealthCheckTimeout = 15
 	}
 
-	// set default port ranges
-	if config.StartPort == 0 {
-		// default to 5800
-		config.StartPort = 5800
-	} else if config.StartPort < 1 {
+	if config.StartPort < 1 {
 		return Config{}, fmt.Errorf("startPort must be greater than 1")
 	}
 
@@ -134,7 +220,30 @@ func LoadConfigFromReader(r io.Reader) (Config, error) {
 		}
 	}
 
-	// iterate over the models and replace any ${PORT} with the next available port
+	/* check macro constraint rules:
+
+	- name must fit the regex ^[a-zA-Z0-9_-]+$
+	- names must be less than 64 characters (no reason, just cause)
+	- name can not be any reserved macros: PORT
+	- macro values must be less than 1024 characters
+	*/
+	macroNameRegex := regexp.MustCompile(`^[a-zA-Z0-9_-]+$`)
+	for macroName, macroValue := range config.Macros {
+		if len(macroName) >= 64 {
+			return Config{}, fmt.Errorf("macro name '%s' exceeds maximum length of 63 characters", macroName)
+		}
+		if !macroNameRegex.MatchString(macroName) {
+			return Config{}, fmt.Errorf("macro name '%s' contains invalid characters, must match pattern ^[a-zA-Z0-9_-]+$", macroName)
+		}
+		if len(macroValue) >= 1024 {
+			return Config{}, fmt.Errorf("macro value for '%s' exceeds maximum length of 1024 characters", macroName)
+		}
+		switch macroName {
+		case "PORT":
+			return Config{}, fmt.Errorf("macro name '%s' is reserved and cannot be used", macroName)
+		}
+	}
+
 	// Get and sort all model IDs first, makes testing more consistent
 	modelIds := make([]string, 0, len(config.Models))
 	for modelId := range config.Models {
@@ -142,23 +251,59 @@ func LoadConfigFromReader(r io.Reader) (Config, error) {
 	}
 	sort.Strings(modelIds) // This guarantees stable iteration order
 
-	// iterate over the sorted models
 	nextPort := config.StartPort
 	for _, modelId := range modelIds {
 		modelConfig := config.Models[modelId]
-		if strings.Contains(modelConfig.Cmd, "${PORT}") {
-			modelConfig.Cmd = strings.ReplaceAll(modelConfig.Cmd, "${PORT}", strconv.Itoa(nextPort))
-			if modelConfig.Proxy == "" {
-				modelConfig.Proxy = fmt.Sprintf("http://localhost:%d", nextPort)
-			} else {
-				modelConfig.Proxy = strings.ReplaceAll(modelConfig.Proxy, "${PORT}", strconv.Itoa(nextPort))
-			}
-			nextPort++
-			config.Models[modelId] = modelConfig
-		} else if modelConfig.Proxy == "" {
-			return Config{}, fmt.Errorf("model %s requires a proxy value when not using automatic ${PORT}", modelId)
+
+		// go through model config fields: cmd, cmdStop, proxy, checkEndPoint and replace macros with macro values
+		for macroName, macroValue := range config.Macros {
+			macroSlug := fmt.Sprintf("${%s}", macroName)
+			modelConfig.Cmd = strings.ReplaceAll(modelConfig.Cmd, macroSlug, macroValue)
+			modelConfig.CmdStop = strings.ReplaceAll(modelConfig.CmdStop, macroSlug, macroValue)
+			modelConfig.Proxy = strings.ReplaceAll(modelConfig.Proxy, macroSlug, macroValue)
+			modelConfig.CheckEndpoint = strings.ReplaceAll(modelConfig.CheckEndpoint, macroSlug, macroValue)
+			modelConfig.Filters.StripParams = strings.ReplaceAll(modelConfig.Filters.StripParams, macroSlug, macroValue)
 		}
+
+		// enforce ${PORT} used in both cmd and proxy
+		if !strings.Contains(modelConfig.Cmd, "${PORT}") && strings.Contains(modelConfig.Proxy, "${PORT}") {
+			return Config{}, fmt.Errorf("model %s: proxy uses ${PORT} but cmd does not - ${PORT} is only available when used in cmd", modelId)
+		}
+
+		// only iterate over models that use ${PORT} to keep port numbers from increasing unnecessarily
+		if strings.Contains(modelConfig.Cmd, "${PORT}") || strings.Contains(modelConfig.Proxy, "${PORT}") || strings.Contains(modelConfig.CmdStop, "${PORT}") {
+			nextPortStr := strconv.Itoa(nextPort)
+			modelConfig.Cmd = strings.ReplaceAll(modelConfig.Cmd, "${PORT}", nextPortStr)
+			modelConfig.CmdStop = strings.ReplaceAll(modelConfig.CmdStop, "${PORT}", nextPortStr)
+			modelConfig.Proxy = strings.ReplaceAll(modelConfig.Proxy, "${PORT}", nextPortStr)
+			nextPort++
+		}
+
+		// make sure there are no unknown macros that have not been replaced
+		macroPattern := regexp.MustCompile(`\$\{([a-zA-Z0-9_-]+)\}`)
+		fieldMap := map[string]string{
+			"cmd":           modelConfig.Cmd,
+			"cmdStop":       modelConfig.CmdStop,
+			"proxy":         modelConfig.Proxy,
+			"checkEndpoint": modelConfig.CheckEndpoint,
+		}
+
+		for fieldName, fieldValue := range fieldMap {
+			matches := macroPattern.FindAllStringSubmatch(fieldValue, -1)
+			for _, match := range matches {
+				macroName := match[1]
+				if macroName == "PID" && fieldName == "cmdStop" {
+					continue // this is ok, has to be replaced by process later
+				}
+				if _, exists := config.Macros[macroName]; !exists {
+					return Config{}, fmt.Errorf("unknown macro '${%s}' found in %s.%s", macroName, modelId, fieldName)
+				}
+			}
+		}
+
+		config.Models[modelId] = modelConfig
 	}
+
 	config = AddDefaultGroupToConfig(config)
 	// check that members are all unique in the groups
 	memberUsage := make(map[string]string) // maps member to group it appears in
@@ -228,14 +373,30 @@ func AddDefaultGroupToConfig(config Config) Config {
 }
 
 func SanitizeCommand(cmdStr string) ([]string, error) {
-	// Remove trailing backslashes
-	cmdStr = strings.ReplaceAll(cmdStr, "\\ \n", " ")
-	cmdStr = strings.ReplaceAll(cmdStr, "\\\n", " ")
+	var cleanedLines []string
+	for _, line := range strings.Split(cmdStr, "\n") {
+		trimmed := strings.TrimSpace(line)
+		// Skip comment lines
+		if strings.HasPrefix(trimmed, "#") {
+			continue
+		}
+		// Handle trailing backslashes by replacing with space
+		if strings.HasSuffix(trimmed, "\\") {
+			cleanedLines = append(cleanedLines, strings.TrimSuffix(trimmed, "\\")+" ")
+		} else {
+			cleanedLines = append(cleanedLines, line)
+		}
+	}
+
+	// put it back together
+	cmdStr = strings.Join(cleanedLines, "\n")
 
 	// Split the command into arguments
-	args, err := shlex.Split(cmdStr)
-	if err != nil {
-		return nil, err
+	var args []string
+	if runtime.GOOS == "windows" {
+		args = shlex.Windows.Split(cmdStr)
+	} else {
+		args = shlex.Posix.Split(cmdStr)
 	}
 
 	// Ensure the command is not empty

proxy/config_posix_test.go

@@ -0,0 +1,233 @@
+//go:build !windows
+
+package proxy
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestConfig_SanitizeCommand(t *testing.T) {
+	// Test a command with spaces and newlines
+	args, err := SanitizeCommand(`python model1.py \
+		-a "double quotes" \
+		--arg2 'single quotes'
+		-s
+		# comment 1
+		--arg3 123 \
+
+		  # comment 2
+		--arg4 '"string in string"'
+
+
+		# this will get stripped out as well as the white space above
+		-c "'single quoted'"
+		`)
+	assert.NoError(t, err)
+	assert.Equal(t, []string{
+		"python", "model1.py",
+		"-a", "double quotes",
+		"--arg2", "single quotes",
+		"-s",
+		"--arg3", "123",
+		"--arg4", `"string in string"`,
+		"-c", `'single quoted'`,
+	}, args)
+
+	// Test an empty command
+	args, err = SanitizeCommand("")
+	assert.Error(t, err)
+	assert.Nil(t, args)
+}
+
+// Test the default values are automatically set for global, model and group configurations
+// after loading the configuration
+func TestConfig_DefaultValuesPosix(t *testing.T) {
+	content := `
+models:
+  model1:
+    cmd: path/to/cmd --port ${PORT}
+`
+
+	config, err := LoadConfigFromReader(strings.NewReader(content))
+	assert.NoError(t, err)
+	assert.Equal(t, 120, config.HealthCheckTimeout)
+	assert.Equal(t, 5800, config.StartPort)
+	assert.Equal(t, "info", config.LogLevel)
+
+	// Test default group exists
+	defaultGroup, exists := config.Groups["(default)"]
+	assert.True(t, exists, "default group should exist")
+	if assert.NotNil(t, defaultGroup, "default group should not be nil") {
+		assert.Equal(t, true, defaultGroup.Swap)
+		assert.Equal(t, true, defaultGroup.Exclusive)
+		assert.Equal(t, false, defaultGroup.Persistent)
+		assert.Equal(t, []string{"model1"}, defaultGroup.Members)
+	}
+
+	model1, exists := config.Models["model1"]
+	assert.True(t, exists, "model1 should exist")
+	if assert.NotNil(t, model1, "model1 should not be nil") {
+		assert.Equal(t, "path/to/cmd --port 5800", model1.Cmd) // has the port replaced
+		assert.Equal(t, "", model1.CmdStop)
+		assert.Equal(t, "http://localhost:5800", model1.Proxy)
+		assert.Equal(t, "/health", model1.CheckEndpoint)
+		assert.Equal(t, []string{}, model1.Aliases)
+		assert.Equal(t, []string{}, model1.Env)
+		assert.Equal(t, 0, model1.UnloadAfter)
+		assert.Equal(t, false, model1.Unlisted)
+		assert.Equal(t, "", model1.UseModelName)
+		assert.Equal(t, 0, model1.ConcurrencyLimit)
+	}
+
+	// default empty filter exists
+	assert.Equal(t, "", model1.Filters.StripParams)
+}
+
+func TestConfig_LoadPosix(t *testing.T) {
+	// Create a temporary YAML file for testing
+	tempDir, err := os.MkdirTemp("", "test-config")
+	if err != nil {
+		t.Fatalf("Failed to create temporary directory: %v", err)
+	}
+	defer os.RemoveAll(tempDir)
+
+	tempFile := filepath.Join(tempDir, "config.yaml")
+	content := `
+macros:
+  svr-path: "path/to/server"
+models:
+  model1:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8080"
+    name: "Model 1"
+    description: "This is model 1"
+    aliases:
+      - "m1"
+      - "model-one"
+    env:
+      - "VAR1=value1"
+      - "VAR2=value2"
+    checkEndpoint: "/health"
+  model2:
+    cmd: ${svr-path} --arg1 one
+    proxy: "http://localhost:8081"
+    aliases:
+      - "m2"
+    checkEndpoint: "/"
+  model3:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8081"
+    aliases:
+      - "mthree"
+    checkEndpoint: "/"
+  model4:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8082"
+    checkEndpoint: "/"
+
+healthCheckTimeout: 15
+profiles:
+  test:
+    - model1
+    - model2
+groups:
+  group1:
+    swap: true
+    exclusive: false
+    members: ["model2"]
+  forever:
+    exclusive: false
+    persistent: true
+    members:
+      - "model4"
+`
+
+	if err := os.WriteFile(tempFile, []byte(content), 0644); err != nil {
+		t.Fatalf("Failed to write temporary file: %v", err)
+	}
+
+	// Load the config and verify
+	config, err := LoadConfig(tempFile)
+	if err != nil {
+		t.Fatalf("Failed to load config: %v", err)
+	}
+
+	expected := Config{
+		LogLevel:  "info",
+		StartPort: 5800,
+		Macros: map[string]string{
+			"svr-path": "path/to/server",
+		},
+		Models: map[string]ModelConfig{
+			"model1": {
+				Cmd:           "path/to/cmd --arg1 one",
+				Proxy:         "http://localhost:8080",
+				Aliases:       []string{"m1", "model-one"},
+				Env:           []string{"VAR1=value1", "VAR2=value2"},
+				CheckEndpoint: "/health",
+				Name:          "Model 1",
+				Description:   "This is model 1",
+			},
+			"model2": {
+				Cmd:           "path/to/server --arg1 one",
+				Proxy:         "http://localhost:8081",
+				Aliases:       []string{"m2"},
+				Env:           []string{},
+				CheckEndpoint: "/",
+			},
+			"model3": {
+				Cmd:           "path/to/cmd --arg1 one",
+				Proxy:         "http://localhost:8081",
+				Aliases:       []string{"mthree"},
+				Env:           []string{},
+				CheckEndpoint: "/",
+			},
+			"model4": {
+				Cmd:           "path/to/cmd --arg1 one",
+				Proxy:         "http://localhost:8082",
+				CheckEndpoint: "/",
+				Aliases:       []string{},
+				Env:           []string{},
+			},
+		},
+		HealthCheckTimeout: 15,
+		Profiles: map[string][]string{
+			"test": {"model1", "model2"},
+		},
+		aliases: map[string]string{
+			"m1":        "model1",
+			"model-one": "model1",
+			"m2":        "model2",
+			"mthree":    "model3",
+		},
+		Groups: map[string]GroupConfig{
+			DEFAULT_GROUP_ID: {
+				Swap:      true,
+				Exclusive: true,
+				Members:   []string{"model1", "model3"},
+			},
+			"group1": {
+				Swap:      true,
+				Exclusive: false,
+				Members:   []string{"model2"},
+			},
+			"forever": {
+				Swap:       true,
+				Exclusive:  false,
+				Persistent: true,
+				Members:    []string{"model4"},
+			},
+		},
+	}
+
+	assert.Equal(t, expected, config)
+
+	realname, found := config.RealModelName("m1")
+	assert.True(t, found)
+	assert.Equal(t, "model1", realname)
+}

proxy/config_test.go

@@ -1,146 +1,12 @@
 package proxy
 
 import (
-	"os"
-	"path/filepath"
 	"strings"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
 )
 
-func TestConfig_Load(t *testing.T) {
-	// Create a temporary YAML file for testing
-	tempDir, err := os.MkdirTemp("", "test-config")
-	if err != nil {
-		t.Fatalf("Failed to create temporary directory: %v", err)
-	}
-	defer os.RemoveAll(tempDir)
-
-	tempFile := filepath.Join(tempDir, "config.yaml")
-	content := `
-models:
-  model1:
-    cmd: path/to/cmd --arg1 one
-    proxy: "http://localhost:8080"
-    aliases:
-      - "m1"
-      - "model-one"
-    env:
-      - "VAR1=value1"
-      - "VAR2=value2"
-    checkEndpoint: "/health"
-  model2:
-    cmd: path/to/cmd --arg1 one
-    proxy: "http://localhost:8081"
-    aliases:
-      - "m2"
-    checkEndpoint: "/"
-  model3:
-    cmd: path/to/cmd --arg1 one
-    proxy: "http://localhost:8081"
-    aliases:
-      - "mthree"
-    checkEndpoint: "/"
-  model4:
-    cmd: path/to/cmd --arg1 one
-    proxy: "http://localhost:8082"
-    checkEndpoint: "/"
-
-healthCheckTimeout: 15
-profiles:
-  test:
-    - model1
-    - model2
-groups:
-  group1:
-    swap: true
-    exclusive: false
-    members: ["model2"]
-  forever:
-    exclusive: false
-    persistent: true
-    members:
-      - "model4"
-`
-
-	if err := os.WriteFile(tempFile, []byte(content), 0644); err != nil {
-		t.Fatalf("Failed to write temporary file: %v", err)
-	}
-
-	// Load the config and verify
-	config, err := LoadConfig(tempFile)
-	if err != nil {
-		t.Fatalf("Failed to load config: %v", err)
-	}
-
-	expected := Config{
-		StartPort: 5800,
-		Models: map[string]ModelConfig{
-			"model1": {
-				Cmd:           "path/to/cmd --arg1 one",
-				Proxy:         "http://localhost:8080",
-				Aliases:       []string{"m1", "model-one"},
-				Env:           []string{"VAR1=value1", "VAR2=value2"},
-				CheckEndpoint: "/health",
-			},
-			"model2": {
-				Cmd:           "path/to/cmd --arg1 one",
-				Proxy:         "http://localhost:8081",
-				Aliases:       []string{"m2"},
-				Env:           nil,
-				CheckEndpoint: "/",
-			},
-			"model3": {
-				Cmd:           "path/to/cmd --arg1 one",
-				Proxy:         "http://localhost:8081",
-				Aliases:       []string{"mthree"},
-				Env:           nil,
-				CheckEndpoint: "/",
-			},
-			"model4": {
-				Cmd:           "path/to/cmd --arg1 one",
-				Proxy:         "http://localhost:8082",
-				CheckEndpoint: "/",
-			},
-		},
-		HealthCheckTimeout: 15,
-		Profiles: map[string][]string{
-			"test": {"model1", "model2"},
-		},
-		aliases: map[string]string{
-			"m1":        "model1",
-			"model-one": "model1",
-			"m2":        "model2",
-			"mthree":    "model3",
-		},
-		Groups: map[string]GroupConfig{
-			DEFAULT_GROUP_ID: {
-				Swap:      true,
-				Exclusive: true,
-				Members:   []string{"model1", "model3"},
-			},
-			"group1": {
-				Swap:      true,
-				Exclusive: false,
-				Members:   []string{"model2"},
-			},
-			"forever": {
-				Swap:       true,
-				Exclusive:  false,
-				Persistent: true,
-				Members:    []string{"model4"},
-			},
-		},
-	}
-
-	assert.Equal(t, expected, config)
-
-	realname, found := config.RealModelName("m1")
-	assert.True(t, found)
-	assert.Equal(t, "model1", realname)
-}
-
 func TestConfig_GroupMemberIsUnique(t *testing.T) {
 	content := `
 models:
@@ -258,34 +124,6 @@ func TestConfig_FindConfig(t *testing.T) {
 	assert.Equal(t, ModelConfig{}, modelConfig)
 }
 
-func TestConfig_SanitizeCommand(t *testing.T) {
-
-	// Test a command with spaces and newlines
-	args, err := SanitizeCommand(`python model1.py \
-    -a "double quotes" \
-    --arg2 'single quotes'
-	-s
-	--arg3 123 \
-	--arg4 '"string in string"'
-	-c "'single quoted'"
-	`)
-	assert.NoError(t, err)
-	assert.Equal(t, []string{
-		"python", "model1.py",
-		"-a", "double quotes",
-		"--arg2", "single quotes",
-		"-s",
-		"--arg3", "123",
-		"--arg4", `"string in string"`,
-		"-c", `'single quoted'`,
-	}, args)
-
-	// Test an empty command
-	args, err = SanitizeCommand("")
-	assert.Error(t, err)
-	assert.Nil(t, args)
-}
-
 func TestConfig_AutomaticPortAssignments(t *testing.T) {
 
 	t.Run("Default Port Ranges", func(t *testing.T) {
@@ -356,6 +194,134 @@ models:
     cmd: svr --port 111
 `
 		_, err := LoadConfigFromReader(strings.NewReader(content))
-		assert.Equal(t, "model model1 requires a proxy value when not using automatic ${PORT}", err.Error())
+		assert.Equal(t, "model model1: proxy uses ${PORT} but cmd does not - ${PORT} is only available when used in cmd", err.Error())
 	})
 }
+
+func TestConfig_MacroReplacement(t *testing.T) {
+	content := `
+startPort: 9990
+macros:
+  svr-path: "path/to/server"
+  argOne: "--arg1"
+  argTwo: "--arg2"
+  autoPort: "--port ${PORT}"
+
+models:
+  model1:
+    cmd: |
+      ${svr-path} ${argTwo}
+      # the automatic ${PORT} is replaced
+      ${autoPort}
+      ${argOne}
+      --arg3 three
+    cmdStop: |
+      /path/to/stop.sh --port ${PORT} ${argTwo}
+`
+
+	config, err := LoadConfigFromReader(strings.NewReader(content))
+	assert.NoError(t, err)
+	sanitizedCmd, err := SanitizeCommand(config.Models["model1"].Cmd)
+	assert.NoError(t, err)
+	assert.Equal(t, "path/to/server --arg2 --port 9990 --arg1 --arg3 three", strings.Join(sanitizedCmd, " "))
+
+	sanitizedCmdStop, err := SanitizeCommand(config.Models["model1"].CmdStop)
+	assert.NoError(t, err)
+	assert.Equal(t, "/path/to/stop.sh --port 9990 --arg2", strings.Join(sanitizedCmdStop, " "))
+}
+
+func TestConfig_MacroErrorOnUnknownMacros(t *testing.T) {
+	tests := []struct {
+		name    string
+		field   string
+		content string
+	}{
+		{
+			name:  "unknown macro in cmd",
+			field: "cmd",
+			content: `
+startPort: 9990
+macros:
+  svr-path: "path/to/server"
+models:
+  model1:
+    cmd: |
+      ${svr-path} --port ${PORT}
+      ${unknownMacro}
+`,
+		},
+		{
+			name:  "unknown macro in cmdStop",
+			field: "cmdStop",
+			content: `
+startPort: 9990
+macros:
+  svr-path: "path/to/server"
+models:
+  model1:
+    cmd: "${svr-path} --port ${PORT}"
+    cmdStop: "kill ${unknownMacro}"
+`,
+		},
+		{
+			name:  "unknown macro in proxy",
+			field: "proxy",
+			content: `
+startPort: 9990
+macros:
+  svr-path: "path/to/server"
+models:
+  model1:
+    cmd: "${svr-path} --port ${PORT}"
+    proxy: "http://localhost:${unknownMacro}"
+`,
+		},
+		{
+			name:  "unknown macro in checkEndpoint",
+			field: "checkEndpoint",
+			content: `
+startPort: 9990
+macros:
+  svr-path: "path/to/server"
+models:
+  model1:
+    cmd: "${svr-path} --port ${PORT}"
+    checkEndpoint: "http://localhost:${unknownMacro}/health"
+`,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			_, err := LoadConfigFromReader(strings.NewReader(tt.content))
+			assert.Error(t, err)
+			assert.Contains(t, err.Error(), "unknown macro '${unknownMacro}' found in model1."+tt.field)
+			//t.Log(err)
+		})
+	}
+}
+
+func TestConfig_ModelFilters(t *testing.T) {
+	content := `
+macros:
+  default_strip: "temperature, top_p"
+models:
+  model1:
+    cmd: path/to/cmd --port ${PORT}
+    filters:
+      strip_params: "model, top_k, ${default_strip}, , ,"
+`
+	config, err := LoadConfigFromReader(strings.NewReader(content))
+	assert.NoError(t, err)
+	modelConfig, ok := config.Models["model1"]
+	if !assert.True(t, ok) {
+		t.FailNow()
+	}
+
+	// make sure `model` and enmpty strings are not in the list
+	assert.Equal(t, "model, top_k, temperature, top_p, , ,", modelConfig.Filters.StripParams)
+	sanitized, err := modelConfig.Filters.SanitizedStripParams()
+	if assert.NoError(t, err) {
+		assert.Equal(t, []string{"temperature", "top_k", "top_p"}, sanitized)
+	}
+}

proxy/config_windows_test.go

@@ -0,0 +1,230 @@
+//go:build windows
+
+package proxy
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestConfig_SanitizeCommand(t *testing.T) {
+	// does not support single quoted strings like in config_posix_test.go
+	args, err := SanitizeCommand(`python model1.py \
+
+	-a "double quotes" \
+	-s
+	--arg3 123 \
+
+	   # comment 2
+	--arg4 '"string in string"'
+
+
+
+	# this will get stripped out as well as the white space above
+	-c "'single quoted'"
+	`)
+	assert.NoError(t, err)
+	assert.Equal(t, []string{
+		"python", "model1.py",
+		"-a", "double quotes",
+		"-s",
+		"--arg3", "123",
+		"--arg4", "'string in string'", // this is a little weird but the lexer says so...?
+		"-c", `'single quoted'`,
+	}, args)
+
+	// Test an empty command
+	args, err = SanitizeCommand("")
+	assert.Error(t, err)
+	assert.Nil(t, args)
+}
+
+func TestConfig_DefaultValuesWindows(t *testing.T) {
+	content := `
+models:
+  model1:
+    cmd: path/to/cmd --port ${PORT}
+`
+
+	config, err := LoadConfigFromReader(strings.NewReader(content))
+	assert.NoError(t, err)
+	assert.Equal(t, 120, config.HealthCheckTimeout)
+	assert.Equal(t, 5800, config.StartPort)
+	assert.Equal(t, "info", config.LogLevel)
+
+	// Test default group exists
+	defaultGroup, exists := config.Groups["(default)"]
+	assert.True(t, exists, "default group should exist")
+	if assert.NotNil(t, defaultGroup, "default group should not be nil") {
+		assert.Equal(t, true, defaultGroup.Swap)
+		assert.Equal(t, true, defaultGroup.Exclusive)
+		assert.Equal(t, false, defaultGroup.Persistent)
+		assert.Equal(t, []string{"model1"}, defaultGroup.Members)
+	}
+
+	model1, exists := config.Models["model1"]
+	assert.True(t, exists, "model1 should exist")
+	if assert.NotNil(t, model1, "model1 should not be nil") {
+		assert.Equal(t, "path/to/cmd --port 5800", model1.Cmd) // has the port replaced
+		assert.Equal(t, "taskkill /f /t /pid ${PID}", model1.CmdStop)
+		assert.Equal(t, "http://localhost:5800", model1.Proxy)
+		assert.Equal(t, "/health", model1.CheckEndpoint)
+		assert.Equal(t, []string{}, model1.Aliases)
+		assert.Equal(t, []string{}, model1.Env)
+		assert.Equal(t, 0, model1.UnloadAfter)
+		assert.Equal(t, false, model1.Unlisted)
+		assert.Equal(t, "", model1.UseModelName)
+		assert.Equal(t, 0, model1.ConcurrencyLimit)
+	}
+
+	// default empty filter exists
+	assert.Equal(t, "", model1.Filters.StripParams)
+}
+
+func TestConfig_LoadWindows(t *testing.T) {
+	// Create a temporary YAML file for testing
+	tempDir, err := os.MkdirTemp("", "test-config")
+	if err != nil {
+		t.Fatalf("Failed to create temporary directory: %v", err)
+	}
+	defer os.RemoveAll(tempDir)
+
+	tempFile := filepath.Join(tempDir, "config.yaml")
+	content := `
+macros:
+  svr-path: "path/to/server"
+models:
+  model1:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8080"
+    aliases:
+      - "m1"
+      - "model-one"
+    env:
+      - "VAR1=value1"
+      - "VAR2=value2"
+    checkEndpoint: "/health"
+  model2:
+    cmd: ${svr-path} --arg1 one
+    proxy: "http://localhost:8081"
+    aliases:
+      - "m2"
+    checkEndpoint: "/"
+  model3:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8081"
+    aliases:
+      - "mthree"
+    checkEndpoint: "/"
+  model4:
+    cmd: path/to/cmd --arg1 one
+    proxy: "http://localhost:8082"
+    checkEndpoint: "/"
+
+healthCheckTimeout: 15
+profiles:
+  test:
+    - model1
+    - model2
+groups:
+  group1:
+    swap: true
+    exclusive: false
+    members: ["model2"]
+  forever:
+    exclusive: false
+    persistent: true
+    members:
+      - "model4"
+`
+
+	if err := os.WriteFile(tempFile, []byte(content), 0644); err != nil {
+		t.Fatalf("Failed to write temporary file: %v", err)
+	}
+
+	// Load the config and verify
+	config, err := LoadConfig(tempFile)
+	if err != nil {
+		t.Fatalf("Failed to load config: %v", err)
+	}
+
+	expected := Config{
+		LogLevel:  "info",
+		StartPort: 5800,
+		Macros: map[string]string{
+			"svr-path": "path/to/server",
+		},
+		Models: map[string]ModelConfig{
+			"model1": {
+				Cmd:           "path/to/cmd --arg1 one",
+				CmdStop:       "taskkill /f /t /pid ${PID}",
+				Proxy:         "http://localhost:8080",
+				Aliases:       []string{"m1", "model-one"},
+				Env:           []string{"VAR1=value1", "VAR2=value2"},
+				CheckEndpoint: "/health",
+			},
+			"model2": {
+				Cmd:           "path/to/server --arg1 one",
+				CmdStop:       "taskkill /f /t /pid ${PID}",
+				Proxy:         "http://localhost:8081",
+				Aliases:       []string{"m2"},
+				Env:           []string{},
+				CheckEndpoint: "/",
+			},
+			"model3": {
+				Cmd:           "path/to/cmd --arg1 one",
+				CmdStop:       "taskkill /f /t /pid ${PID}",
+				Proxy:         "http://localhost:8081",
+				Aliases:       []string{"mthree"},
+				Env:           []string{},
+				CheckEndpoint: "/",
+			},
+			"model4": {
+				Cmd:           "path/to/cmd --arg1 one",
+				CmdStop:       "taskkill /f /t /pid ${PID}",
+				Proxy:         "http://localhost:8082",
+				CheckEndpoint: "/",
+				Aliases:       []string{},
+				Env:           []string{},
+			},
+		},
+		HealthCheckTimeout: 15,
+		Profiles: map[string][]string{
+			"test": {"model1", "model2"},
+		},
+		aliases: map[string]string{
+			"m1":        "model1",
+			"model-one": "model1",
+			"m2":        "model2",
+			"mthree":    "model3",
+		},
+		Groups: map[string]GroupConfig{
+			DEFAULT_GROUP_ID: {
+				Swap:      true,
+				Exclusive: true,
+				Members:   []string{"model1", "model3"},
+			},
+			"group1": {
+				Swap:      true,
+				Exclusive: false,
+				Members:   []string{"model2"},
+			},
+			"forever": {
+				Swap:       true,
+				Exclusive:  false,
+				Persistent: true,
+				Members:    []string{"model4"},
+			},
+		},
+	}
+
+	assert.Equal(t, expected, config)
+
+	realname, found := config.RealModelName("m1")
+	assert.True(t, found)
+	assert.Equal(t, "model1", realname)
+}

proxy/helpers_test.go

@@ -9,6 +9,7 @@ import (
 	"testing"
 
 	"github.com/gin-gonic/gin"
+	"gopkg.in/yaml.v3"
 )
 
 var (
@@ -45,26 +46,41 @@ func TestMain(m *testing.M) {
 func getSimpleResponderPath() string {
 	goos := runtime.GOOS
 	goarch := runtime.GOARCH
-	return filepath.Join("..", "build", fmt.Sprintf("simple-responder_%s_%s", goos, goarch))
+
+	if goos == "windows" {
+		return filepath.Join("..", "build", "simple-responder.exe")
+	} else {
+		return filepath.Join("..", "build", fmt.Sprintf("simple-responder_%s_%s", goos, goarch))
+	}
 }
 
-func getTestSimpleResponderConfig(expectedMessage string) ModelConfig {
+func getTestPort() int {
 	portMutex.Lock()
 	defer portMutex.Unlock()
 
 	port := nextTestPort
 	nextTestPort++
 
-	return getTestSimpleResponderConfigPort(expectedMessage, port)
+	return port
+}
+
+func getTestSimpleResponderConfig(expectedMessage string) ModelConfig {
+	return getTestSimpleResponderConfigPort(expectedMessage, getTestPort())
 }
 
 func getTestSimpleResponderConfigPort(expectedMessage string, port int) ModelConfig {
 	binaryPath := getSimpleResponderPath()
 
-	// Create a process configuration
-	return ModelConfig{
-		Cmd:           fmt.Sprintf("%s --port %d --silent --respond %s", binaryPath, port, expectedMessage),
-		Proxy:         fmt.Sprintf("http://127.0.0.1:%d", port),
-		CheckEndpoint: "/health",
+	// Create a YAML string with just the values we want to set
+	yamlStr := fmt.Sprintf(`
+cmd: '%s --port %d --silent --respond %s'
+proxy: "http://127.0.0.1:%d"
+`, binaryPath, port, expectedMessage, port)
+
+	var cfg ModelConfig
+	if err := yaml.Unmarshal([]byte(yamlStr), &cfg); err != nil {
+		panic(fmt.Sprintf("failed to unmarshal test config: %v in [%s]", err, yamlStr))
 	}
+
+	return cfg
 }

proxy/html/index.html

@@ -1,14 +0,0 @@
-<!DOCTYPE html>
-<html lang="en">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>llama-swap</title>
-</head>
-<body>
-    <h1>llama-swap</h1>
-    <p>
-        <a href="/logs">view logs</a> | <a href="/upstream">configured models</a> | <a href="https://github.com/mostlygeek/llama-swap">github</a>
-    </p>
-</body>
-</html>

proxy/html/logs.html

@@ -1,259 +0,0 @@
-<!DOCTYPE html>
-<html lang="en">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>Logs</title>
-    <style>
-        body {
-            margin: 0;
-            height: 100vh;
-            display: flex;
-            flex-direction: column;
-            font-family: "Courier New", Courier, monospace;
-        }
-        .log-container {
-            display: flex;
-            flex: 1;
-            gap: 0.5em;
-            margin: 0.5em;
-            min-height: 0;
-        }
-        .log-column {
-            display: flex;
-            flex-direction: column;
-            flex: 1;
-            min-width: 0;
-            transition: flex 0.3s ease;
-        }
-        .log-column.minimized {
-            flex: 0.1;
-            max-width: 50px;
-            border: 1px solid #777;
-            color: green;
-        }
-        .log-controls {
-            display: grid;
-            grid-template-columns: 1fr auto;
-            gap: 0.5em;
-            margin-bottom: 0.5em;
-        }
-        .log-controls input {
-            width: 100%;
-            padding: 4px;
-        }
-        .log-controls input:focus {
-           outline: none;
-        }
-        .log-stream {
-            flex: 1;
-            padding: 1em;
-            background: #f4f4f4;
-            overflow-y: auto;
-            white-space: pre-wrap;
-            word-wrap: break-word;
-            min-height: 0;
-        }
-
-        .regex-error {
-            background-color: #ff0000 !important;
-        }
-
-        /* Make headers clickable and show pointer cursor */
-        h2 {
-            cursor: pointer;
-            user-select: none;
-            margin: 0 0 0.5em 0;
-            padding: 0.5em;
-        }
-
-        h2:hover {
-            background-color: rgba(0, 0, 0, 0.05);
-        }
-
-        /* Dark mode styles */
-        @media (prefers-color-scheme: dark) {
-            body {
-                background-color: #333;
-                color: #fff;
-            }
-
-            .log-stream {
-                background: #444;
-                color: #fff;
-            }
-
-            .log-controls input {
-                background: #555;
-                color: #fff;
-                border: 1px solid #777;
-            }
-
-            .log-controls button {
-                background: #555;
-                color: #fff;
-                border: 1px solid #777;
-            }
-
-            h2:hover {
-                background-color: rgba(255, 255, 255, 0.1);
-            }
-        }
-
-        /* Hide content when minimized */
-        .log-column.minimized .log-controls,
-        .log-column.minimized .log-stream {
-            display: none;
-        }
-
-        .log-column.minimized h2 {
-            writing-mode: vertical-rl;
-            text-orientation: mixed;
-            transform: rotate(180deg);
-            white-space: nowrap;
-            margin: auto;
-        }
-    </style>
-</head>
-<body>
-    <div class="log-container">
-        <div class="log-column">
-            <h2>Proxy Logs</h2>
-            <div class="log-controls">
-                <input type="text" id="proxy-filter-input" placeholder="proxy regex filter">
-                <button id="proxy-clear-button">clear</button>
-            </div>
-            <pre class="log-stream" id="proxy-log-stream">Waiting for proxy logs...</pre>
-        </div>
-        <div class="log-column minimized">
-            <h2>Upstream Logs</h2>
-            <div class="log-controls">
-                <input type="text" id="upstream-filter-input" placeholder="upstream regex filter">
-                <button id="upstream-clear-button">clear</button>
-            </div>
-            <pre class="log-stream" id="upstream-log-stream">Waiting for upstream logs...</pre>
-        </div>
-    </div>
-    <script>
-        class LogStream {
-            constructor(streamElement, filterInput, clearButton, endpoint) {
-                this.streamElement = streamElement;
-                this.filterInput = filterInput;
-                this.clearButton = clearButton;
-                this.endpoint = endpoint;
-                this.logData = "";
-                this.regexFilter = null;
-                this.eventSource = null;
-
-                this.initialize();
-            }
-
-            initialize() {
-                this.filterInput.addEventListener('input', () => this.updateFilter());
-                this.clearButton.addEventListener('click', () => {
-                    this.filterInput.value = "";
-                    this.regexFilter = null;
-                    this.render();
-                });
-                this.setupEventSource();
-            }
-
-            setupEventSource() {
-                if (typeof(EventSource) === "undefined") {
-                    this.logData = "SSE Not supported by this browser.";
-                    this.render();
-                    return;
-                }
-
-                const connect = () => {
-                    this.eventSource = new EventSource(this.endpoint);
-
-                    this.eventSource.onmessage = (event) => {
-                        this.logData += event.data;
-                        this.logData = this.logData.slice(-1024 * 100);
-                        this.render();
-                    };
-
-                    this.eventSource.onerror = (err) => {
-                        // Close the current connection
-                        this.eventSource.close();
-
-                        this.logData += "\nConnection lost. Retrying in 5 seconds...\n";
-                        this.render();
-
-                        // Attempt to reconnect after 5 seconds
-                        setTimeout(() => {
-                            this.logData += "Attempting to reconnect...\n";
-                            this.render();
-                            connect();
-                        }, 5000);
-                    };
-                };
-
-                // Initial connection
-                connect();
-            }
-
-            render() {
-                let content = this.logData;
-
-                if (this.regexFilter) {
-                    const lines = content.split('\n');
-                    const filteredLines = lines.filter(line => this.regexFilter.test(line));
-                    content = filteredLines.length > 0 ? filteredLines.join('\n') + '\n' : "";
-                }
-
-                this.streamElement.textContent = content;
-                this.streamElement.scrollTop = this.streamElement.scrollHeight;
-            }
-
-            updateFilter() {
-                const pattern = this.filterInput.value.trim();
-                this.filterInput.classList.remove('regex-error');
-
-                if (!pattern) {
-                    this.regexFilter = null;
-                    this.render();
-                    return;
-                }
-
-                try {
-                    this.regexFilter = new RegExp(pattern);
-                } catch (e) {
-                    console.error("Invalid regex pattern:", e);
-                    this.regexFilter = null;
-                    this.filterInput.classList.add('regex-error');
-                    return;
-                }
-
-                this.render();
-            }
-        }
-
-        // Initialize both log streams
-        document.addEventListener('DOMContentLoaded', () => {
-            new LogStream(
-                document.getElementById('proxy-log-stream'),
-                document.getElementById('proxy-filter-input'),
-                document.getElementById('proxy-clear-button'),
-                "/logs/streamSSE/proxy"
-            );
-
-            new LogStream(
-                document.getElementById('upstream-log-stream'),
-                document.getElementById('upstream-filter-input'),
-                document.getElementById('upstream-clear-button'),
-                "/logs/streamSSE/upstream"
-            );
-
-            // Initialize clickable headers
-            document.querySelectorAll('h2').forEach(header => {
-                header.addEventListener('click', () => {
-                    const column = header.closest('.log-column');
-                    column.classList.toggle('minimized');
-                });
-            });
-        });
-    </script>
-</body>
-</html>

proxy/html_files.go

@@ -1,10 +0,0 @@
-package proxy
-
-import "embed"
-
-//go:embed html
-var htmlFiles embed.FS
-
-func getHTMLFile(path string) ([]byte, error) {
-	return htmlFiles.ReadFile("html/" + path)
-}

proxy/process.go

@@ -23,9 +23,6 @@ const (
 	StateReady    ProcessState = ProcessState("ready")
 	StateStopping ProcessState = ProcessState("stopping")
 
-	// failed a health check on start and will not be recovered
-	StateFailed ProcessState = ProcessState("failed")
-
 	// process is shutdown and will not be restarted
 	StateShutdown ProcessState = ProcessState("shutdown")
 )
@@ -42,8 +39,11 @@ type Process struct {
 	config ModelConfig
 	cmd    *exec.Cmd
 
-	// for p.cmd.Wait() select { ... }
-	cmdWaitChan chan error
+	// PR #155 called to cancel the upstream process
+	cancelUpstream context.CancelFunc
+
+	// closed when command exits
+	cmdWaitChan chan struct{}
 
 	processLogger *LogMonitor
 	proxyLogger   *LogMonitor
@@ -61,37 +61,40 @@ type Process struct {
 	// used to block on multiple start() calls
 	waitStarting sync.WaitGroup
 
-	// for managing shutdown state
-	shutdownCtx    context.Context
-	shutdownCancel context.CancelFunc
-
 	// for managing concurrency limits
 	concurrencyLimitSemaphore chan struct{}
+
+	// used for testing to override the default value
+	gracefulStopTimeout time.Duration
+
+	// track the number of failed starts
+	failedStartCount int
 }
 
 func NewProcess(ID string, healthCheckTimeout int, config ModelConfig, processLogger *LogMonitor, proxyLogger *LogMonitor) *Process {
-	ctx, cancel := context.WithCancel(context.Background())
 	concurrentLimit := 10
 	if config.ConcurrencyLimit > 0 {
 		concurrentLimit = config.ConcurrencyLimit
-	} else {
-		proxyLogger.Debugf("Concurrency limit for model %s not set, defaulting to 10", ID)
 	}
+
 	return &Process{
 		ID:                      ID,
 		config:                  config,
 		cmd:                     nil,
-		cmdWaitChan:             make(chan error, 1),
+		cancelUpstream:          nil,
 		processLogger:           processLogger,
 		proxyLogger:             proxyLogger,
 		healthCheckTimeout:      healthCheckTimeout,
 		healthCheckLoopInterval: 5 * time.Second, /* default, can not be set by user - used for testing */
 		state:                   StateStopped,
-		shutdownCtx:             ctx,
-		shutdownCancel:          cancel,
 
 		// concurrency limit
 		concurrencyLimitSemaphore: make(chan struct{}, concurrentLimit),
+
+		// To be removed when migration over exec.CommandContext is complete
+		// stop timeout
+		gracefulStopTimeout: 10 * time.Second,
+		cmdWaitChan:         make(chan struct{}),
 	}
 }
 
@@ -133,12 +136,12 @@ func isValidTransition(from, to ProcessState) bool {
 	case StateStopped:
 		return to == StateStarting
 	case StateStarting:
-		return to == StateReady || to == StateFailed || to == StateStopping
+		return to == StateReady || to == StateStopping || to == StateStopped
 	case StateReady:
 		return to == StateStopping
 	case StateStopping:
 		return to == StateStopped || to == StateShutdown
-	case StateFailed, StateShutdown:
+	case StateShutdown:
 		return false // No transitions allowed from these states
 	}
 	return false
@@ -185,17 +188,26 @@ func (p *Process) start() error {
 
 	p.waitStarting.Add(1)
 	defer p.waitStarting.Done()
+	cmdContext, ctxCancelUpstream := context.WithCancel(context.Background())
 
-	p.cmd = exec.Command(args[0], args[1:]...)
+	p.cmd = exec.CommandContext(cmdContext, args[0], args[1:]...)
 	p.cmd.Stdout = p.processLogger
 	p.cmd.Stderr = p.processLogger
-	p.cmd.Env = p.config.Env
+	p.cmd.Env = append(p.cmd.Environ(), p.config.Env...)
+	p.cmd.Cancel = p.cmdStopUpstreamProcess
+	p.cmd.WaitDelay = p.gracefulStopTimeout
+	p.cancelUpstream = ctxCancelUpstream
+	p.cmdWaitChan = make(chan struct{})
 
+	p.failedStartCount++ // this will be reset to zero when the process has successfully started
+
+	p.proxyLogger.Debugf("<%s> Executing start command: %s, env: %s", p.ID, strings.Join(args, " "), strings.Join(p.config.Env, ", "))
 	err = p.cmd.Start()
 
 	// Set process state to failed
 	if err != nil {
-		if curState, swapErr := p.swapState(StateStarting, StateFailed); swapErr != nil {
+		if curState, swapErr := p.swapState(StateStarting, StateStopped); swapErr != nil {
+			p.state = StateStopped // force it into a stopped state
 			return fmt.Errorf(
 				"failed to start command and state swap failed. command error: %v, current state: %v, state swap error: %v",
 				err, curState, swapErr,
@@ -205,11 +217,7 @@ func (p *Process) start() error {
 	}
 
 	// Capture the exit error for later signalling
-	go func() {
-		exitErr := p.cmd.Wait()
-		p.proxyLogger.Debugf("<%s> cmd.Wait() returned error: %v", p.ID, exitErr)
-		p.cmdWaitChan <- exitErr
-	}()
+	go p.waitForCmd()
 
 	// One of three things can happen at this stage:
 	// 1. The command exits unexpectedly
@@ -225,67 +233,38 @@ func (p *Process) start() error {
 
 	// a "none" means don't check for health ... I could have picked a better word :facepalm:
 	if checkEndpoint != "none" {
-		// keep default behaviour
-		if checkEndpoint == "" {
-			checkEndpoint = "/health"
-		}
-
 		proxyTo := p.config.Proxy
 		healthURL, err := url.JoinPath(proxyTo, checkEndpoint)
 		if err != nil {
 			return fmt.Errorf("failed to create health check URL proxy=%s and checkEndpoint=%s", proxyTo, checkEndpoint)
 		}
 
-		checkDeadline, cancelHealthCheck := context.WithDeadline(
-			context.Background(),
-			checkStartTime.Add(maxDuration),
-		)
-		defer cancelHealthCheck()
-
-	loop:
 		// Ready Check loop
 		for {
-			select {
-			case <-checkDeadline.Done():
-				if curState, err := p.swapState(StateStarting, StateFailed); err != nil {
-					return fmt.Errorf("health check timed out after %vs AND state swap failed: %v, current state: %v", maxDuration.Seconds(), err, curState)
-				} else {
-					return fmt.Errorf("health check timed out after %vs", maxDuration.Seconds())
+			currentState := p.CurrentState()
+			if currentState != StateStarting {
+				if currentState == StateStopped {
+					return fmt.Errorf("upstream command exited prematurely but successfully")
 				}
-			case <-p.shutdownCtx.Done():
 				return errors.New("health check interrupted due to shutdown")
-			case exitErr := <-p.cmdWaitChan:
-				if exitErr != nil {
-					p.proxyLogger.Warnf("<%s> upstream command exited prematurely with error: %v", p.ID, exitErr)
-					if curState, err := p.swapState(StateStarting, StateFailed); err != nil {
-						return fmt.Errorf("upstream command exited unexpectedly: %s AND state swap failed: %v, current state: %v", exitErr.Error(), err, curState)
-					} else {
-						return fmt.Errorf("upstream command exited unexpectedly: %s", exitErr.Error())
-					}
-				} else {
-					p.proxyLogger.Warnf("<%s> upstream command exited prematurely but successfully", p.ID)
-					if curState, err := p.swapState(StateStarting, StateFailed); err != nil {
-						return fmt.Errorf("upstream command exited prematurely but successfully AND state swap failed: %v, current state: %v", err, curState)
-					} else {
-						return fmt.Errorf("upstream command exited prematurely but successfully")
-					}
-				}
-			default:
-				if err := p.checkHealthEndpoint(healthURL); err == nil {
-					p.proxyLogger.Infof("<%s> Health check passed on %s", p.ID, healthURL)
-					cancelHealthCheck()
-					break loop
-				} else {
-					if strings.Contains(err.Error(), "connection refused") {
-						endTime, _ := checkDeadline.Deadline()
-						ttl := time.Until(endTime)
-						p.proxyLogger.Debugf("<%s> Connection refused on %s, giving up in %.0fs (normal during startup)", p.ID, healthURL, ttl.Seconds())
-					} else {
-						p.proxyLogger.Debugf("<%s> Health check error on %s, %v (normal during startup)", p.ID, healthURL, err)
-					}
-				}
 			}
 
+			if time.Since(checkStartTime) > maxDuration {
+				p.stopCommand()
+				return fmt.Errorf("health check timed out after %vs", maxDuration.Seconds())
+			}
+
+			if err := p.checkHealthEndpoint(healthURL); err == nil {
+				p.proxyLogger.Infof("<%s> Health check passed on %s", p.ID, healthURL)
+				break
+			} else {
+				if strings.Contains(err.Error(), "connection refused") {
+					ttl := time.Until(checkStartTime.Add(maxDuration))
+					p.proxyLogger.Debugf("<%s> Connection refused on %s, giving up in %.0fs (normal during startup)", p.ID, healthURL, ttl.Seconds())
+				} else {
+					p.proxyLogger.Debugf("<%s> Health check error on %s, %v (normal during startup)", p.ID, healthURL, err)
+				}
+			}
 			<-time.After(p.healthCheckLoopInterval)
 		}
 	}
@@ -316,6 +295,7 @@ func (p *Process) start() error {
 	if curState, err := p.swapState(StateStarting, StateReady); err != nil {
 		return fmt.Errorf("failed to set Process state to ready: current state: %v, error: %v", curState, err)
 	} else {
+		p.failedStartCount = 0
 		return nil
 	}
 }
@@ -339,20 +319,13 @@ func (p *Process) StopImmediately() {
 		return
 	}
 
-	p.proxyLogger.Debugf("<%s> Stopping process", p.ID)
-
-	// calling Stop() when state is invalid is a no-op
+	p.proxyLogger.Debugf("<%s> Stopping process, current state: %s", p.ID, p.CurrentState())
 	if curState, err := p.swapState(StateReady, StateStopping); err != nil {
 		p.proxyLogger.Infof("<%s> Stop() Ready -> StateStopping err: %v, current state: %v", p.ID, err, curState)
 		return
 	}
 
-	// stop the process with a graceful exit timeout
-	p.stopCommand(5 * time.Second)
-
-	if curState, err := p.swapState(StateStopping, StateStopped); err != nil {
-		p.proxyLogger.Infof("<%s> Stop() StateStopping -> StateStopped err: %v, current state: %v", p.ID, err, curState)
-	}
+	p.stopCommand()
 }
 
 // Shutdown is called when llama-swap is shutting down. It will give a little bit
@@ -360,58 +333,30 @@ func (p *Process) StopImmediately() {
 // is in the state of starting, it will cancel it and shut it down. Once a process is in
 // the StateShutdown state, it can not be started again.
 func (p *Process) Shutdown() {
-	p.shutdownCancel()
-	p.stopCommand(5 * time.Second)
+	if !isValidTransition(p.CurrentState(), StateStopping) {
+		return
+	}
+
+	p.stopCommand()
+	// just force it to this state since there is no recovery from shutdown
 	p.state = StateShutdown
 }
 
 // stopCommand will send a SIGTERM to the process and wait for it to exit.
 // If it does not exit within 5 seconds, it will send a SIGKILL.
-func (p *Process) stopCommand(sigtermTTL time.Duration) {
+func (p *Process) stopCommand() {
 	stopStartTime := time.Now()
 	defer func() {
 		p.proxyLogger.Debugf("<%s> stopCommand took %v", p.ID, time.Since(stopStartTime))
 	}()
 
-	sigtermTimeout, cancelTimeout := context.WithTimeout(context.Background(), sigtermTTL)
-	defer cancelTimeout()
-
-	if p.cmd == nil || p.cmd.Process == nil {
-		p.proxyLogger.Debugf("<%s> cmd or cmd.Process is nil (normal during config reload)", p.ID)
+	if p.cancelUpstream == nil {
+		p.proxyLogger.Errorf("<%s> stopCommand has a nil p.cancelUpstream()", p.ID)
 		return
 	}
 
-	if err := p.terminateProcess(); err != nil {
-		p.proxyLogger.Debugf("<%s> Process already terminated: %v (normal during shutdown)", p.ID, err)
-	}
-
-	select {
-	case <-sigtermTimeout.Done():
-		p.proxyLogger.Debugf("<%s> Process timed out waiting to stop, sending KILL signal (normal during shutdown)", p.ID)
-		if err := p.cmd.Process.Kill(); err != nil {
-			p.proxyLogger.Errorf("<%s> Failed to kill process: %v", p.ID, err)
-		}
-	case err := <-p.cmdWaitChan:
-		// Note: in start(), p.cmdWaitChan also has a select { ... }. That should be OK
-		// because if we make it here then the cmd has been successfully running and made it
-		// through the health check. There is a possibility that the cmd crashed after the health check
-		// succeeded but that's not a case llama-swap is handling for now.
-		if err != nil {
-			if errno, ok := err.(syscall.Errno); ok {
-				p.proxyLogger.Errorf("<%s> errno >> %v", p.ID, errno)
-			} else if exitError, ok := err.(*exec.ExitError); ok {
-				if strings.Contains(exitError.String(), "signal: terminated") {
-					p.proxyLogger.Debugf("<%s> Process stopped OK", p.ID)
-				} else if strings.Contains(exitError.String(), "signal: interrupt") {
-					p.proxyLogger.Debugf("<%s> Process interrupted OK", p.ID)
-				} else {
-					p.proxyLogger.Warnf("<%s> ExitError >> %v, exit code: %d", p.ID, exitError, exitError.ExitCode())
-				}
-			} else {
-				p.proxyLogger.Errorf("<%s> Process exited >> %v", p.ID, err)
-			}
-		}
-	}
+	p.cancelUpstream()
+	<-p.cmdWaitChan
 }
 
 func (p *Process) checkHealthEndpoint(healthURL string) error {
@@ -444,7 +389,7 @@ func (p *Process) ProxyRequest(w http.ResponseWriter, r *http.Request) {
 
 	// prevent new requests from being made while stopping or irrecoverable
 	currentState := p.CurrentState()
-	if currentState == StateFailed || currentState == StateShutdown || currentState == StateStopping {
+	if currentState == StateShutdown || currentState == StateStopping {
 		http.Error(w, fmt.Sprintf("Process can not ProxyRequest, state is %s", currentState), http.StatusServiceUnavailable)
 		return
 	}
@@ -526,3 +471,79 @@ func (p *Process) ProxyRequest(w http.ResponseWriter, r *http.Request) {
 	p.proxyLogger.Debugf("<%s> request %s - start: %v, total: %v",
 		p.ID, r.RequestURI, startDuration, totalTime)
 }
+
+// waitForCmd waits for the command to exit and handles exit conditions depending on current state
+func (p *Process) waitForCmd() {
+	exitErr := p.cmd.Wait()
+	p.proxyLogger.Debugf("<%s> cmd.Wait() returned error: %v", p.ID, exitErr)
+
+	if exitErr != nil {
+		if errno, ok := exitErr.(syscall.Errno); ok {
+			p.proxyLogger.Errorf("<%s> errno >> %v", p.ID, errno)
+		} else if exitError, ok := exitErr.(*exec.ExitError); ok {
+			if strings.Contains(exitError.String(), "signal: terminated") {
+				p.proxyLogger.Debugf("<%s> Process stopped OK", p.ID)
+			} else if strings.Contains(exitError.String(), "signal: interrupt") {
+				p.proxyLogger.Debugf("<%s> Process interrupted OK", p.ID)
+			} else {
+				p.proxyLogger.Warnf("<%s> ExitError >> %v, exit code: %d", p.ID, exitError, exitError.ExitCode())
+			}
+		} else {
+			if exitErr.Error() != "context canceled" /* this is normal */ {
+				p.proxyLogger.Errorf("<%s> Process exited >> %v", p.ID, exitErr)
+			}
+		}
+	}
+
+	currentState := p.CurrentState()
+	switch currentState {
+	case StateStopping:
+		if curState, err := p.swapState(StateStopping, StateStopped); err != nil {
+			p.proxyLogger.Errorf("<%s> Process exited but could not swap to StateStopped. curState=%s, err: %v", p.ID, curState, err)
+			p.state = StateStopped
+		}
+	default:
+		p.proxyLogger.Infof("<%s> process exited but not StateStopping, current state: %s", p.ID, currentState)
+		p.state = StateStopped // force it to be in this state
+	}
+	close(p.cmdWaitChan)
+}
+
+// cmdStopUpstreamProcess attemps to stop the upstream process gracefully
+func (p *Process) cmdStopUpstreamProcess() error {
+	p.processLogger.Debugf("<%s> cmdStopUpstreamProcess() initiating graceful stop of upstream process", p.ID)
+
+	// this should never happen ...
+	if p.cmd == nil || p.cmd.Process == nil {
+		p.proxyLogger.Debugf("<%s> cmd or cmd.Process is nil (normal during config reload)", p.ID)
+		return fmt.Errorf("<%s> process is nil or cmd is nil, skipping graceful stop", p.ID)
+	}
+
+	if p.config.CmdStop != "" {
+		// replace ${PID} with the pid of the process
+		stopArgs, err := SanitizeCommand(strings.ReplaceAll(p.config.CmdStop, "${PID}", fmt.Sprintf("%d", p.cmd.Process.Pid)))
+		if err != nil {
+			p.proxyLogger.Errorf("<%s> Failed to sanitize stop command: %v", p.ID, err)
+			return err
+		}
+
+		p.proxyLogger.Debugf("<%s> Executing stop command: %s", p.ID, strings.Join(stopArgs, " "))
+
+		stopCmd := exec.Command(stopArgs[0], stopArgs[1:]...)
+		stopCmd.Stdout = p.processLogger
+		stopCmd.Stderr = p.processLogger
+		stopCmd.Env = p.cmd.Env
+
+		if err := stopCmd.Run(); err != nil {
+			p.proxyLogger.Errorf("<%s> Failed to exec stop command: %v", p.ID, err)
+			return err
+		}
+	} else {
+		if err := p.cmd.Process.Signal(syscall.SIGTERM); err != nil {
+			p.proxyLogger.Errorf("<%s> Failed to send SIGTERM to process: %v", p.ID, err)
+			return err
+		}
+	}
+
+	return nil
+}

proxy/process_stop.go

@@ -1,9 +0,0 @@
-//go:build !windows
-
-package proxy
-
-import "syscall"
-
-func (p *Process) terminateProcess() error {
-	return p.cmd.Process.Signal(syscall.SIGTERM)
-}

proxy/process_stop_windows.go

@@ -1,14 +0,0 @@
-//go:build windows
-
-package proxy
-
-import (
-	"fmt"
-	"os/exec"
-)
-
-func (p *Process) terminateProcess() error {
-	pid := fmt.Sprintf("%d", p.cmd.Process.Pid)
-	cmd := exec.Command("taskkill", "/f", "/t", "/pid", pid)
-	return cmd.Run()
-}

proxy/process_test.go

@@ -5,6 +5,7 @@ import (
 	"net/http"
 	"net/http/httptest"
 	"os"
+	"runtime"
 	"sync"
 	"testing"
 	"time"
@@ -105,8 +106,8 @@ func TestProcess_BrokenModelConfig(t *testing.T) {
 
 	w = httptest.NewRecorder()
 	process.ProxyRequest(w, req)
-	assert.Equal(t, http.StatusServiceUnavailable, w.Code)
-	assert.Contains(t, w.Body.String(), "Process can not ProxyRequest, state is failed")
+	assert.Equal(t, http.StatusBadGateway, w.Code)
+	assert.Contains(t, w.Body.String(), "start() failed: ")
 }
 
 func TestProcess_UnloadAfterTTL(t *testing.T) {
@@ -247,18 +248,14 @@ func TestProcess_SwapState(t *testing.T) {
 	}{
 		{"Stopped to Starting", StateStopped, StateStopped, StateStarting, nil, StateStarting},
 		{"Starting to Ready", StateStarting, StateStarting, StateReady, nil, StateReady},
-		{"Starting to Failed", StateStarting, StateStarting, StateFailed, nil, StateFailed},
 		{"Starting to Stopping", StateStarting, StateStarting, StateStopping, nil, StateStopping},
+		{"Starting to Stopped", StateStarting, StateStarting, StateStopped, nil, StateStopped},
 		{"Ready to Stopping", StateReady, StateReady, StateStopping, nil, StateStopping},
 		{"Stopping to Stopped", StateStopping, StateStopping, StateStopped, nil, StateStopped},
 		{"Stopping to Shutdown", StateStopping, StateStopping, StateShutdown, nil, StateShutdown},
 		{"Stopped to Ready", StateStopped, StateStopped, StateReady, ErrInvalidStateTransition, StateStopped},
-		{"Starting to Stopped", StateStarting, StateStarting, StateStopped, ErrInvalidStateTransition, StateStarting},
 		{"Ready to Starting", StateReady, StateReady, StateStarting, ErrInvalidStateTransition, StateReady},
-		{"Ready to Failed", StateReady, StateReady, StateFailed, ErrInvalidStateTransition, StateReady},
 		{"Stopping to Ready", StateStopping, StateStopping, StateReady, ErrInvalidStateTransition, StateStopping},
-		{"Failed to Stopped", StateFailed, StateFailed, StateStopped, ErrInvalidStateTransition, StateFailed},
-		{"Failed to Starting", StateFailed, StateFailed, StateStarting, ErrInvalidStateTransition, StateFailed},
 		{"Shutdown to Stopped", StateShutdown, StateShutdown, StateStopped, ErrInvalidStateTransition, StateShutdown},
 		{"Shutdown to Starting", StateShutdown, StateShutdown, StateStarting, ErrInvalidStateTransition, StateShutdown},
 		{"Expected state mismatch", StateStopped, StateStarting, StateStarting, ErrExpectedStateMismatch, StateStopped},
@@ -338,7 +335,7 @@ func TestProcess_ExitInterruptsHealthCheck(t *testing.T) {
 	process.healthCheckLoopInterval = time.Second // make it faster
 	err := process.start()
 	assert.Equal(t, "upstream command exited prematurely but successfully", err.Error())
-	assert.Equal(t, process.CurrentState(), StateFailed)
+	assert.Equal(t, process.CurrentState(), StateStopped)
 }
 
 func TestProcess_ConcurrencyLimit(t *testing.T) {
@@ -393,3 +390,104 @@ func TestProcess_StopImmediately(t *testing.T) {
 	process.StopImmediately()
 	assert.Equal(t, process.CurrentState(), StateStopped)
 }
+
+// Test that SIGKILL is sent when gracefulStopTimeout is reached and properly terminates
+// the upstream command
+func TestProcess_ForceStopWithKill(t *testing.T) {
+	if runtime.GOOS == "windows" {
+		t.Skip("skipping SIGTERM test on Windows ")
+	}
+
+	expectedMessage := "test_sigkill"
+	binaryPath := getSimpleResponderPath()
+	port := getTestPort()
+
+	config := ModelConfig{
+		// note --ignore-sig-term which ignores the SIGTERM signal so a SIGKILL must be sent
+		// to force the process to exit
+		Cmd:           fmt.Sprintf("%s --port %d --respond %s --silent --ignore-sig-term", binaryPath, port, expectedMessage),
+		Proxy:         fmt.Sprintf("http://127.0.0.1:%d", port),
+		CheckEndpoint: "/health",
+	}
+
+	process := NewProcess("stop_immediate", 2, config, debugLogger, debugLogger)
+	defer process.Stop()
+
+	// reduce to make testing go faster
+	process.gracefulStopTimeout = time.Second
+
+	err := process.start()
+	assert.Nil(t, err)
+	assert.Equal(t, process.CurrentState(), StateReady)
+
+	waitChan := make(chan struct{})
+	go func() {
+		// slow, but will get killed by StopImmediate
+		req := httptest.NewRequest("GET", "/slow-respond?echo=12345&delay=2s", nil)
+		w := httptest.NewRecorder()
+		process.ProxyRequest(w, req)
+
+		// StatusOK because that was already sent before the kill
+		assert.Equal(t, http.StatusOK, w.Code)
+
+		// unexpected EOF because the kill happened, the "1" is sent before the kill
+		// then the unexpected EOF is sent after the kill
+		if runtime.GOOS == "windows" {
+			assert.Contains(t, w.Body.String(), "wsarecv: An existing connection was forcibly closed by the remote host")
+		} else {
+			assert.Contains(t, w.Body.String(), "unexpected EOF")
+		}
+
+		close(waitChan)
+	}()
+
+	<-time.After(time.Millisecond)
+	process.StopImmediately()
+	assert.Equal(t, process.CurrentState(), StateStopped)
+
+	// the request should have been interrupted by SIGKILL
+	<-waitChan
+}
+
+func TestProcess_StopCmd(t *testing.T) {
+	config := getTestSimpleResponderConfig("test_stop_cmd")
+
+	if runtime.GOOS == "windows" {
+		config.CmdStop = "taskkill /f /t /pid ${PID}"
+	} else {
+		config.CmdStop = "kill -TERM ${PID}"
+	}
+
+	process := NewProcess("testStopCmd", 2, config, debugLogger, debugLogger)
+	defer process.Stop()
+
+	err := process.start()
+	assert.Nil(t, err)
+	assert.Equal(t, process.CurrentState(), StateReady)
+	process.StopImmediately()
+	assert.Equal(t, process.CurrentState(), StateStopped)
+}
+
+func TestProcess_EnvironmentSetCorrectly(t *testing.T) {
+	expectedMessage := "test_env_not_emptied"
+	config := getTestSimpleResponderConfig(expectedMessage)
+
+	// ensure that the the default config does not blank out the inherited environment
+	configWEnv := config
+
+	// ensure the additiona variables are appended to the process' environment
+	configWEnv.Env = append(configWEnv.Env, "TEST_ENV1=1", "TEST_ENV2=2")
+
+	process1 := NewProcess("env_test", 2, config, debugLogger, debugLogger)
+	process2 := NewProcess("env_test", 2, configWEnv, debugLogger, debugLogger)
+
+	process1.start()
+	defer process1.Stop()
+	process2.start()
+	defer process2.Stop()
+
+	assert.NotZero(t, len(process1.cmd.Environ()))
+	assert.NotZero(t, len(process2.cmd.Environ()))
+	assert.Equal(t, len(process1.cmd.Environ())+2, len(process2.cmd.Environ()), "process2 should have 2 more environment variables than process1")
+
+}

proxy/proxymanager.go

@@ -2,13 +2,11 @@ package proxy
 
 import (
 	"bytes"
-	"encoding/json"
 	"fmt"
 	"io"
 	"mime/multipart"
 	"net/http"
 	"os"
-	"sort"
 	"strconv"
 	"strings"
 	"sync"
@@ -163,38 +161,59 @@ func (pm *ProxyManager) setupGinEngine() {
 	pm.ginEngine.GET("/logs/stream/:logMonitorID", pm.streamLogsHandler)
 	pm.ginEngine.GET("/logs/streamSSE/:logMonitorID", pm.streamLogsHandlerSSE)
 
-	pm.ginEngine.GET("/upstream", pm.upstreamIndex)
+	/**
+	 * User Interface Endpoints
+	 */
+	pm.ginEngine.GET("/", func(c *gin.Context) {
+		c.Redirect(http.StatusFound, "/ui")
+	})
+
+	pm.ginEngine.GET("/upstream", func(c *gin.Context) {
+		c.Redirect(http.StatusFound, "/ui/models")
+	})
 	pm.ginEngine.Any("/upstream/:model_id/*upstreamPath", pm.proxyToUpstream)
 
 	pm.ginEngine.GET("/unload", pm.unloadAllModelsHandler)
-
 	pm.ginEngine.GET("/running", pm.listRunningProcessesHandler)
 
-	pm.ginEngine.GET("/", func(c *gin.Context) {
-		// Set the Content-Type header to text/html
-		c.Header("Content-Type", "text/html")
-
-		// Write the embedded HTML content to the response
-		htmlData, err := getHTMLFile("index.html")
-		if err != nil {
-			c.String(http.StatusInternalServerError, err.Error())
-			return
-		}
-		_, err = c.Writer.Write(htmlData)
-		if err != nil {
-			c.String(http.StatusInternalServerError, fmt.Sprintf("failed to write response: %v", err))
-			return
-		}
-	})
-
 	pm.ginEngine.GET("/favicon.ico", func(c *gin.Context) {
-		if data, err := getHTMLFile("favicon.ico"); err == nil {
+		if data, err := reactStaticFS.ReadFile("ui_dist/favicon.ico"); err == nil {
 			c.Data(http.StatusOK, "image/x-icon", data)
 		} else {
 			c.String(http.StatusInternalServerError, err.Error())
 		}
 	})
 
+	reactFS, err := GetReactFS()
+	if err != nil {
+		pm.proxyLogger.Errorf("Failed to load React filesystem: %v", err)
+	} else {
+
+		// serve files that exist under /ui/*
+		pm.ginEngine.StaticFS("/ui", reactFS)
+
+		// server SPA for UI under /ui/*
+		pm.ginEngine.NoRoute(func(c *gin.Context) {
+			if !strings.HasPrefix(c.Request.URL.Path, "/ui") {
+				c.AbortWithStatus(http.StatusNotFound)
+				return
+			}
+
+			file, err := reactFS.Open("index.html")
+			if err != nil {
+				c.String(http.StatusInternalServerError, err.Error())
+				return
+			}
+			defer file.Close()
+			http.ServeContent(c.Writer, c.Request, "index.html", time.Now(), file)
+
+		})
+	}
+
+	// see: proxymanager_api.go
+	// add API handler functions
+	addApiHandlers(pm)
+
 	// Disable console color for testing
 	gin.DisableConsoleColor()
 }
@@ -269,32 +288,41 @@ func (pm *ProxyManager) swapProcessGroup(requestedModel string) (*ProcessGroup,
 }
 
 func (pm *ProxyManager) listModelsHandler(c *gin.Context) {
-	data := []interface{}{}
+	data := make([]gin.H, 0, len(pm.config.Models))
+	createdTime := time.Now().Unix()
+
 	for id, modelConfig := range pm.config.Models {
 		if modelConfig.Unlisted {
 			continue
 		}
 
-		data = append(data, map[string]interface{}{
+		record := gin.H{
 			"id":       id,
 			"object":   "model",
-			"created":  time.Now().Unix(),
+			"created":  createdTime,
 			"owned_by": "llama-swap",
-		})
+		}
+
+		if name := strings.TrimSpace(modelConfig.Name); name != "" {
+			record["name"] = name
+		}
+		if desc := strings.TrimSpace(modelConfig.Description); desc != "" {
+			record["description"] = desc
+		}
+
+		data = append(data, record)
 	}
 
-	// Set the Content-Type header to application/json
-	c.Header("Content-Type", "application/json")
-
-	if origin := c.Request.Header.Get("Origin"); origin != "" {
+	// Set CORS headers if origin exists
+	if origin := c.GetHeader("Origin"); origin != "" {
 		c.Header("Access-Control-Allow-Origin", origin)
 	}
 
-	// Encode the data as JSON and write it to the response writer
-	if err := json.NewEncoder(c.Writer).Encode(map[string]interface{}{"data": data}); err != nil {
-		pm.sendErrorResponse(c, http.StatusInternalServerError, fmt.Sprintf("error encoding JSON %s", err.Error()))
-		return
-	}
+	// Use gin's JSON method which handles content-type and encoding
+	c.JSON(http.StatusOK, gin.H{
+		"object": "list",
+		"data":   data,
+	})
 }
 
 func (pm *ProxyManager) proxyToUpstream(c *gin.Context) {
@@ -316,31 +344,6 @@ func (pm *ProxyManager) proxyToUpstream(c *gin.Context) {
 	processGroup.ProxyRequest(requestedModel, c.Writer, c.Request)
 }
 
-func (pm *ProxyManager) upstreamIndex(c *gin.Context) {
-	var html strings.Builder
-
-	html.WriteString("<!doctype HTML>\n<html><body><h1>Available Models</h1><ul>")
-
-	// Extract keys and sort them
-	var modelIDs []string
-	for modelID, modelConfig := range pm.config.Models {
-		if modelConfig.Unlisted {
-			continue
-		}
-
-		modelIDs = append(modelIDs, modelID)
-	}
-	sort.Strings(modelIDs)
-
-	// Iterate over sorted keys
-	for _, modelID := range modelIDs {
-		html.WriteString(fmt.Sprintf("<li><a href=\"/upstream/%s\">%s</a></li>", modelID, modelID))
-	}
-	html.WriteString("</ul></body></html>")
-	c.Header("Content-Type", "text/html")
-	c.String(http.StatusOK, html.String())
-}
-
 func (pm *ProxyManager) proxyOAIHandler(c *gin.Context) {
 	bodyBytes, err := io.ReadAll(c.Request.Body)
 	if err != nil {
@@ -370,11 +373,27 @@ func (pm *ProxyManager) proxyOAIHandler(c *gin.Context) {
 		}
 	}
 
+	// issue #174 strip parameters from the JSON body
+	stripParams, err := pm.config.Models[realModelName].Filters.SanitizedStripParams()
+	if err != nil { // just log it and continue
+		pm.proxyLogger.Errorf("Error sanitizing strip params string: %s, %s", pm.config.Models[realModelName].Filters.StripParams, err.Error())
+	} else {
+		for _, param := range stripParams {
+			pm.proxyLogger.Debugf("<%s> stripping param: %s", realModelName, param)
+			bodyBytes, err = sjson.DeleteBytes(bodyBytes, param)
+			if err != nil {
+				pm.sendErrorResponse(c, http.StatusInternalServerError, fmt.Sprintf("error deleting parameter %s from request", param))
+				return
+			}
+		}
+	}
+
 	c.Request.Body = io.NopCloser(bytes.NewBuffer(bodyBytes))
 
 	// dechunk it as we already have all the body bytes see issue #11
 	c.Request.Header.Del("transfer-encoding")
-	c.Request.Header.Add("content-length", strconv.Itoa(len(bodyBytes)))
+	c.Request.Header.Set("content-length", strconv.Itoa(len(bodyBytes)))
+	c.Request.ContentLength = int64(len(bodyBytes))
 
 	if err := processGroup.ProxyRequest(realModelName, c.Writer, c.Request); err != nil {
 		pm.sendErrorResponse(c, http.StatusInternalServerError, fmt.Sprintf("error proxying request: %s", err.Error()))

proxy/proxymanager_api.go

@@ -0,0 +1,101 @@
+package proxy
+
+import (
+	"net/http"
+	"sort"
+	"time"
+
+	"github.com/gin-gonic/gin"
+)
+
+type Model struct {
+	Id    string `json:"id"`
+	State string `json:"state"`
+}
+
+func addApiHandlers(pm *ProxyManager) {
+	// Add API endpoints for React to consume
+	apiGroup := pm.ginEngine.Group("/api")
+	{
+		apiGroup.GET("/models", pm.apiListModels)
+		apiGroup.GET("/modelsSSE", pm.apiListModelsSSE)
+		apiGroup.POST("/models/unload", pm.apiUnloadAllModels)
+	}
+}
+
+func (pm *ProxyManager) apiUnloadAllModels(c *gin.Context) {
+	pm.StopProcesses(StopImmediately)
+	c.JSON(http.StatusOK, gin.H{"msg": "ok"})
+}
+
+func (pm *ProxyManager) getModelStatus() []Model {
+	// Extract keys and sort them
+	models := []Model{}
+
+	modelIDs := make([]string, 0, len(pm.config.Models))
+	for modelID := range pm.config.Models {
+		modelIDs = append(modelIDs, modelID)
+	}
+	sort.Strings(modelIDs)
+
+	// Iterate over sorted keys
+	for _, modelID := range modelIDs {
+		// Get process state
+		processGroup := pm.findGroupByModelName(modelID)
+		state := "unknown"
+		if processGroup != nil {
+			process := processGroup.processes[modelID]
+			if process != nil {
+				var stateStr string
+				switch process.CurrentState() {
+				case StateReady:
+					stateStr = "ready"
+				case StateStarting:
+					stateStr = "starting"
+				case StateStopping:
+					stateStr = "stopping"
+				case StateShutdown:
+					stateStr = "shutdown"
+				case StateStopped:
+					stateStr = "stopped"
+				default:
+					stateStr = "unknown"
+				}
+				state = stateStr
+			}
+		}
+		models = append(models, Model{
+			Id:    modelID,
+			State: state,
+		})
+	}
+
+	return models
+}
+
+func (pm *ProxyManager) apiListModels(c *gin.Context) {
+	c.JSON(http.StatusOK, pm.getModelStatus())
+}
+
+// stream the models as a SSE
+func (pm *ProxyManager) apiListModelsSSE(c *gin.Context) {
+	c.Header("Content-Type", "text/event-stream")
+	c.Header("Cache-Control", "no-cache")
+	c.Header("Connection", "keep-alive")
+	c.Header("X-Content-Type-Options", "nosniff")
+
+	notify := c.Request.Context().Done()
+
+	// Stream new events
+	for {
+		select {
+		case <-notify:
+			return
+		default:
+			models := pm.getModelStatus()
+			c.SSEvent("message", models)
+			c.Writer.Flush()
+			<-time.After(1000 * time.Millisecond)
+		}
+	}
+}

proxy/proxymanager_loghandlers.go

@@ -11,20 +11,7 @@ import (
 func (pm *ProxyManager) sendLogsHandlers(c *gin.Context) {
 	accept := c.GetHeader("Accept")
 	if strings.Contains(accept, "text/html") {
-		// Set the Content-Type header to text/html
-		c.Header("Content-Type", "text/html")
-
-		// Write the embedded HTML content to the response
-		logsHTML, err := getHTMLFile("logs.html")
-		if err != nil {
-			c.String(http.StatusInternalServerError, err.Error())
-			return
-		}
-		_, err = c.Writer.Write(logsHTML)
-		if err != nil {
-			c.String(http.StatusInternalServerError, fmt.Sprintf("failed to write response: %v", err))
-			return
-		}
+		c.Redirect(http.StatusFound, "/ui/")
 	} else {
 		c.Header("Content-Type", "text/plain")
 		history := pm.muxLogger.GetHistory()

proxy/proxymanager_test.go

@@ -14,6 +14,7 @@ import (
 	"time"
 
 	"github.com/stretchr/testify/assert"
+	"github.com/tidwall/gjson"
 )
 
 func TestProxyManager_SwapProcessCorrectly(t *testing.T) {
@@ -182,11 +183,20 @@ func TestProxyManager_SwapMultiProcessParallelRequests(t *testing.T) {
 }
 
 func TestProxyManager_ListModelsHandler(t *testing.T) {
+
+	model1Config := getTestSimpleResponderConfig("model1")
+	model1Config.Name = "Model 1"
+	model1Config.Description = "Model 1 description is used for testing"
+
+	model2Config := getTestSimpleResponderConfig("model2")
+	model2Config.Name = "     " // empty whitespace only strings will get ignored
+	model2Config.Description = "  "
+
 	config := Config{
 		HealthCheckTimeout: 15,
 		Models: map[string]ModelConfig{
-			"model1": getTestSimpleResponderConfig("model1"),
-			"model2": getTestSimpleResponderConfig("model2"),
+			"model1": model1Config,
+			"model2": model2Config,
 			"model3": getTestSimpleResponderConfig("model3"),
 		},
 		LogLevel: "error",
@@ -212,6 +222,7 @@ func TestProxyManager_ListModelsHandler(t *testing.T) {
 	var response struct {
 		Data []map[string]interface{} `json:"data"`
 	}
+
 	if err := json.Unmarshal(w.Body.Bytes(), &response); err != nil {
 		t.Fatalf("Failed to parse JSON response: %v", err)
 	}
@@ -226,6 +237,7 @@ func TestProxyManager_ListModelsHandler(t *testing.T) {
 		"model3": {},
 	}
 
+	// make all models
 	for _, model := range response.Data {
 		modelID, ok := model["id"].(string)
 		assert.True(t, ok, "model ID should be a string")
@@ -244,6 +256,21 @@ func TestProxyManager_ListModelsHandler(t *testing.T) {
 		ownedBy, ok := model["owned_by"].(string)
 		assert.True(t, ok, "owned_by should be a string")
 		assert.Equal(t, "llama-swap", ownedBy)
+
+		// check for optional name and description
+		if modelID == "model1" {
+			name, ok := model["name"].(string)
+			assert.True(t, ok, "name should be a string")
+			assert.Equal(t, "Model 1", name)
+			description, ok := model["description"].(string)
+			assert.True(t, ok, "description should be a string")
+			assert.Equal(t, "Model 1 description is used for testing", description)
+		} else {
+			_, exists := model["name"]
+			assert.False(t, exists, "unexpected name field for model: %s", modelID)
+			_, exists = model["description"]
+			assert.False(t, exists, "unexpected description field for model: %s", modelID)
+		}
 	}
 
 	// Ensure all expected models were returned
@@ -339,7 +366,7 @@ func TestProxyManager_RunningEndpoint(t *testing.T) {
 			"model1": getTestSimpleResponderConfig("model1"),
 			"model2": getTestSimpleResponderConfig("model2"),
 		},
-		LogLevel: "debug",
+		LogLevel: "warn",
 	})
 
 	// Define a helper struct to parse the JSON response.
@@ -448,7 +475,6 @@ func TestProxyManager_AudioTranscriptionHandler(t *testing.T) {
 // Test useModelName in configuration sends overrides what is sent to upstream
 func TestProxyManager_UseModelName(t *testing.T) {
 	upstreamModelName := "upstreamModel"
-
 	modelConfig := getTestSimpleResponderConfig(upstreamModelName)
 	modelConfig.UseModelName = upstreamModelName
 
@@ -473,6 +499,12 @@ func TestProxyManager_UseModelName(t *testing.T) {
 		proxy.ServeHTTP(w, req)
 		assert.Equal(t, http.StatusOK, w.Code)
 		assert.Contains(t, w.Body.String(), upstreamModelName)
+
+		// make sure the content length was set correctly
+		// simple-responder will return the content length it got in the response
+		body := w.Body.Bytes()
+		contentLength := int(gjson.GetBytes(body, "h_content_length").Int())
+		assert.Equal(t, len(fmt.Sprintf(`{"model":"%s"}`, upstreamModelName)), contentLength)
 	})
 
 	t.Run("useModelName over rides requested model: /v1/audio/transcriptions", func(t *testing.T) {
@@ -617,3 +649,37 @@ func TestProxyManager_ChatContentLength(t *testing.T) {
 	assert.Equal(t, "81", response["h_content_length"])
 	assert.Equal(t, "model1", response["responseMessage"])
 }
+
+func TestProxyManager_FiltersStripParams(t *testing.T) {
+	modelConfig := getTestSimpleResponderConfig("model1")
+	modelConfig.Filters = ModelFilters{
+		StripParams: "temperature, model, stream",
+	}
+
+	config := AddDefaultGroupToConfig(Config{
+		HealthCheckTimeout: 15,
+		LogLevel:           "error",
+		Models: map[string]ModelConfig{
+			"model1": modelConfig,
+		},
+	})
+
+	proxy := New(config)
+	defer proxy.StopProcesses(StopWaitForInflightRequest)
+	reqBody := `{"model":"model1", "temperature":0.1, "x_param":"123", "y_param":"abc", "stream":true}`
+	req := httptest.NewRequest("POST", "/v1/chat/completions", bytes.NewBufferString(reqBody))
+	w := httptest.NewRecorder()
+
+	proxy.ServeHTTP(w, req)
+	assert.Equal(t, http.StatusOK, w.Code)
+	var response map[string]string
+	assert.NoError(t, json.Unmarshal(w.Body.Bytes(), &response))
+
+	// `temperature` and `stream` are gone but model remains
+	assert.Equal(t, `{"model":"model1", "x_param":"123", "y_param":"abc"}`, response["request_body"])
+
+	// assert.Nil(t, response["temperature"])
+	// assert.Equal(t, "123", response["x_param"])
+	// assert.Equal(t, "abc", response["y_param"])
+	// t.Logf("%v", response)
+}

proxy/ui_embed.go

@@ -0,0 +1,24 @@
+package proxy
+
+import (
+	"embed"
+	"io/fs"
+	"net/http"
+)
+
+//go:embed ui_dist
+var reactStaticFS embed.FS
+
+// GetReactFS returns the embedded React filesystem
+func GetReactFS() (http.FileSystem, error) {
+	subFS, err := fs.Sub(reactStaticFS, "ui_dist")
+	if err != nil {
+		return nil, err
+	}
+	return http.FS(subFS), nil
+}
+
+// GetReactIndexHTML returns the main index.html for the React app
+func GetReactIndexHTML() ([]byte, error) {
+	return reactStaticFS.ReadFile("ui_dist/index.html")
+}

scripts/install.sh

@@ -0,0 +1,213 @@
+#!/bin/sh
+# This script installs llama-swap on Linux.
+# It detects the current operating system architecture and installs the appropriate version of llama-swap.
+
+set -eu
+
+LLAMA_SWAP_DEFAULT_ADDRESS=${LLAMA_SWAP_DEFAULT_ADDRESS:-"127.0.0.1:8080"}
+
+red="$( (/usr/bin/tput bold || :; /usr/bin/tput setaf 1 || :) 2>&-)"
+plain="$( (/usr/bin/tput sgr0 || :) 2>&-)"
+
+status() { echo ">>> $*" >&2; }
+error() { echo "${red}ERROR:${plain} $*"; exit 1; }
+warning() { echo "${red}WARNING:${plain} $*"; }
+
+available() { command -v "$1" >/dev/null; }
+require() {
+    _MISSING=''
+    for TOOL in "$@"; do
+        if ! available "$TOOL"; then
+            _MISSING="$_MISSING $TOOL"
+        fi
+    done
+
+    echo "$_MISSING"
+}
+
+SUDO=
+if [ "$(id -u)" -ne 0 ]; then
+    if ! available sudo; then
+        error "This script requires superuser permissions. Please re-run as root."
+    fi
+
+    SUDO="sudo"
+fi
+
+NEEDS=$(require tee tar python3 mktemp)
+if [ -n "$NEEDS" ]; then
+    status "ERROR: The following tools are required but missing:"
+    for NEED in $NEEDS; do
+        echo "  - $NEED"
+    done
+    exit 1
+fi
+
+[ "$(uname -s)" = "Linux" ] || error 'This script is intended to run on Linux only.'
+
+ARCH=$(uname -m)
+case "$ARCH" in
+    x86_64) ARCH="amd64" ;;
+    aarch64|arm64) ARCH="arm64" ;;
+    *) error "Unsupported architecture: $ARCH" ;;
+esac
+
+IS_WSL2=false
+
+KERN=$(uname -r)
+case "$KERN" in
+    *icrosoft*WSL2 | *icrosoft*wsl2) IS_WSL2=true;;
+    *icrosoft) error "Microsoft WSL1 is not currently supported. Please use WSL2 with 'wsl --set-version <distro> 2'" ;;
+    *) ;;
+esac
+
+download_binary() {
+    ASSET_NAME="linux_$ARCH"
+
+    TMPDIR=$(mktemp -d)
+    trap 'rm -rf "${TMPDIR}"' EXIT INT TERM HUP
+    PYTHON_SCRIPT=$(cat <<EOF
+import os
+import json
+import sys
+import urllib.request
+
+ASSET_NAME = "${ASSET_NAME}"
+
+with urllib.request.urlopen("https://api.github.com/repos/mostlygeek/llama-swap/releases/latest") as resp:
+    data = json.load(resp)
+    for asset in data.get("assets", []):
+        if ASSET_NAME in asset.get("name", ""):
+            url = asset["browser_download_url"]
+            break
+    else:
+        print("ERROR: Matching asset not found.", file=sys.stderr)
+        exit(1)
+
+print("Downloading:", url, file=sys.stderr)
+output_path = os.path.join("${TMPDIR}", "llama-swap.tar.gz")
+urllib.request.urlretrieve(url, output_path)
+print(output_path)
+EOF
+)
+
+    TARFILE=$(python3 -c "$PYTHON_SCRIPT")
+    if [ ! -f "$TARFILE" ]; then
+        error "Failed to download binary."
+    fi
+
+    status "Extracting to /usr/local/bin"
+    $SUDO tar -xzf "$TARFILE" -C /usr/local/bin llama-swap
+}
+download_binary
+
+configure_systemd() {
+    if ! id llama-swap >/dev/null 2>&1; then
+        status "Creating llama-swap user..."
+        $SUDO useradd -r -s /bin/false -U -m -d /usr/share/llama-swap llama-swap
+    fi
+    if getent group render >/dev/null 2>&1; then
+        status "Adding llama-swap user to render group..."
+        $SUDO usermod -a -G render llama-swap
+    fi
+    if getent group video >/dev/null 2>&1; then
+        status "Adding llama-swap user to video group..."
+        $SUDO usermod -a -G video llama-swap
+    fi
+    if getent group docker >/dev/null 2>&1; then
+        status "Adding llama-swap user to docker group..."
+        $SUDO usermod -a -G docker llama-swap
+    fi
+
+    status "Adding current user to llama-swap group..."
+    $SUDO usermod -a -G llama-swap "$(whoami)"
+
+    if [ ! -f "/usr/share/llama-swap/config.yaml" ]; then
+        status "Creating default config.yaml..."
+        cat <<EOF | $SUDO -u llama-swap tee /usr/share/llama-swap/config.yaml >/dev/null
+# default 15s likely to fail for default models due to downloading models
+healthCheckTimeout: 60
+
+models:
+  "qwen2.5":
+    cmd: |
+      docker run
+        --rm
+        -p \${PORT}:8080
+        --name qwen2.5
+      ghcr.io/ggml-org/llama.cpp:server
+        -hf bartowski/Qwen2.5-0.5B-Instruct-GGUF:Q4_K_M
+    cmdStop: docker stop qwen2.5
+
+  "smollm2":
+    cmd: |
+      docker run
+        --rm
+        -p \${PORT}:8080
+        --name smollm2
+      ghcr.io/ggml-org/llama.cpp:server
+        -hf bartowski/SmolLM2-135M-Instruct-GGUF:Q4_K_M
+    cmdStop: docker stop smollm2
+EOF
+    fi
+
+    status "Creating llama-swap systemd service..."
+    cat <<EOF | $SUDO tee /etc/systemd/system/llama-swap.service >/dev/null
+[Unit]
+Description=llama-swap
+After=network.target
+
+[Service]
+User=llama-swap
+Group=llama-swap
+
+# set this to match your environment
+ExecStart=/usr/local/bin/llama-swap --config /usr/share/llama-swap/config.yaml --watch-config -listen ${LLAMA_SWAP_DEFAULT_ADDRESS}
+
+Restart=on-failure
+RestartSec=3
+StartLimitBurst=3
+StartLimitInterval=30
+
+[Install]
+WantedBy=multi-user.target
+EOF
+    SYSTEMCTL_RUNNING="$(systemctl is-system-running || true)"
+    case $SYSTEMCTL_RUNNING in
+        running|degraded)
+            status "Enabling and starting llama-swap service..."
+            $SUDO systemctl daemon-reload
+            $SUDO systemctl enable llama-swap
+
+            start_service() { $SUDO systemctl restart llama-swap; }
+            trap start_service EXIT
+            ;;
+        *)
+            warning "systemd is not running"
+            if [ "$IS_WSL2" = true ]; then
+                warning "see https://learn.microsoft.com/en-us/windows/wsl/systemd#how-to-enable-systemd to enable it"
+            fi
+            ;;
+    esac
+}
+
+if available systemctl; then
+    configure_systemd
+fi
+
+install_success() {
+    status "The llama-swap API is now available at http://${LLAMA_SWAP_DEFAULT_ADDRESS}"
+    status 'Customize the config file at /usr/share/llama-swap/config.yaml.'
+    status 'Install complete.'
+}
+
+# WSL2 only supports GPUs via nvidia passthrough
+# so check for nvidia-smi to determine if GPU is available
+if [ "$IS_WSL2" = true ]; then
+    if available nvidia-smi && [ -n "$(nvidia-smi | grep -o "CUDA Version: [0-9]*\.[0-9]*")" ]; then
+        status "Nvidia GPU detected."
+    fi
+    exit 0
+fi
+
+install_success

scripts/uninstall.sh

@@ -0,0 +1,68 @@
+#!/bin/sh
+# This script uninstalls llama-swap on Linux.
+# It removes the binary, systemd service, config.yaml (optional), and llama-swap user and group.
+
+set -eu
+
+red="$( (/usr/bin/tput bold || :; /usr/bin/tput setaf 1 || :) 2>&-)"
+plain="$( (/usr/bin/tput sgr0 || :) 2>&-)"
+
+status() { echo ">>> $*" >&2; }
+error() { echo "${red}ERROR:${plain} $*"; exit 1; }
+warning() { echo "${red}WARNING:${plain} $*"; }
+
+available() { command -v $1 >/dev/null; }
+
+SUDO=
+if [ "$(id -u)" -ne 0 ]; then
+    if ! available sudo; then
+        error "This script requires superuser permissions. Please re-run as root."
+    fi
+
+    SUDO="sudo"
+fi
+
+configure_systemd() {
+    status "Stopping llama-swap service..."
+    $SUDO systemctl stop llama-swap
+
+    status "Disabling llama-swap service..."
+    $SUDO systemctl disable llama-swap
+}
+if available systemctl; then
+    configure_systemd
+fi
+
+if available llama-swap; then
+    status "Removing llama-swap binary..."
+    $SUDO rm $(which llama-swap)
+fi
+
+if [ -f "/usr/share/llama-swap/config.yaml" ]; then
+    while true; do
+        printf "Delete config.yaml (/usr/share/llama-swap/config.yaml)? [y/N] " >&2
+        read answer
+        case "$answer" in
+            [Yy]* ) 
+                $SUDO rm -r /usr/share/llama-swap
+                break
+                ;;
+            [Nn]* | "" ) 
+                break
+                ;;
+            * ) 
+                echo "Invalid input. Please enter y or n."
+                ;;
+        esac
+    done
+fi
+
+if id llama-swap >/dev/null 2>&1; then
+    status "Removing llama-swap user..."
+    $SUDO userdel llama-swap
+fi
+
+if getent group llama-swap >/dev/null 2>&1; then
+    status "Removing llama-swap group..."
+    $SUDO groupdel llama-swap
+fi

ui/.gitignore

@@ -0,0 +1,25 @@
+.vite
+# Logs
+logs
+*.log
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+pnpm-debug.log*
+lerna-debug.log*
+
+node_modules
+dist
+dist-ssr
+*.local
+
+# Editor directories and files
+.vscode/*
+!.vscode/extensions.json
+.idea
+.DS_Store
+*.suo
+*.ntvs*
+*.njsproj
+*.sln
+*.sw?

ui/README.md

@@ -0,0 +1,54 @@
+# React + TypeScript + Vite
+
+This template provides a minimal setup to get React working in Vite with HMR and some ESLint rules.
+
+Currently, two official plugins are available:
+
+- [@vitejs/plugin-react](https://github.com/vitejs/vite-plugin-react/blob/main/packages/plugin-react) uses [Babel](https://babeljs.io/) for Fast Refresh
+- [@vitejs/plugin-react-swc](https://github.com/vitejs/vite-plugin-react/blob/main/packages/plugin-react-swc) uses [SWC](https://swc.rs/) for Fast Refresh
+
+## Expanding the ESLint configuration
+
+If you are developing a production application, we recommend updating the configuration to enable type-aware lint rules:
+
+```js
+export default tseslint.config({
+  extends: [
+    // Remove ...tseslint.configs.recommended and replace with this
+    ...tseslint.configs.recommendedTypeChecked,
+    // Alternatively, use this for stricter rules
+    ...tseslint.configs.strictTypeChecked,
+    // Optionally, add this for stylistic rules
+    ...tseslint.configs.stylisticTypeChecked,
+  ],
+  languageOptions: {
+    // other options...
+    parserOptions: {
+      project: ['./tsconfig.node.json', './tsconfig.app.json'],
+      tsconfigRootDir: import.meta.dirname,
+    },
+  },
+})
+```
+
+You can also install [eslint-plugin-react-x](https://github.com/Rel1cx/eslint-react/tree/main/packages/plugins/eslint-plugin-react-x) and [eslint-plugin-react-dom](https://github.com/Rel1cx/eslint-react/tree/main/packages/plugins/eslint-plugin-react-dom) for React-specific lint rules:
+
+```js
+// eslint.config.js
+import reactX from 'eslint-plugin-react-x'
+import reactDom from 'eslint-plugin-react-dom'
+
+export default tseslint.config({
+  plugins: {
+    // Add the react-x and react-dom plugins
+    'react-x': reactX,
+    'react-dom': reactDom,
+  },
+  rules: {
+    // other rules...
+    // Enable its recommended typescript rules
+    ...reactX.configs['recommended-typescript'].rules,
+    ...reactDom.configs.recommended.rules,
+  },
+})
+```

ui/eslint.config.js

@@ -0,0 +1,28 @@
+import js from '@eslint/js'
+import globals from 'globals'
+import reactHooks from 'eslint-plugin-react-hooks'
+import reactRefresh from 'eslint-plugin-react-refresh'
+import tseslint from 'typescript-eslint'
+
+export default tseslint.config(
+  { ignores: ['dist'] },
+  {
+    extends: [js.configs.recommended, ...tseslint.configs.recommended],
+    files: ['**/*.{ts,tsx}'],
+    languageOptions: {
+      ecmaVersion: 2020,
+      globals: globals.browser,
+    },
+    plugins: {
+      'react-hooks': reactHooks,
+      'react-refresh': reactRefresh,
+    },
+    rules: {
+      ...reactHooks.configs.recommended.rules,
+      'react-refresh/only-export-components': [
+        'warn',
+        { allowConstantExport: true },
+      ],
+    },
+  },
+)

ui/index.html

@@ -0,0 +1,17 @@
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <link rel="icon" type="image/png" href="/favicon-96x96.png" sizes="96x96" />
+    <link rel="icon" type="image/svg+xml" href="/favicon.svg" />
+    <link rel="shortcut icon" href="/favicon.ico" />
+    <link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png" />
+    <link rel="manifest" href="/site.webmanifest" />
+    <title>llama-swap</title>
+  </head>
+  <body >
+    <div id="root"></div>
+    <script type="module" src="/src/main.tsx"></script>
+  </body>
+</html>

ui/package.json

@@ -0,0 +1,33 @@
+{
+  "name": "ui",
+  "private": true,
+  "version": "0.0.0",
+  "type": "module",
+  "scripts": {
+    "dev": "vite",
+    "build": "tsc -b && vite build --emptyOutDir",
+    "lint": "eslint .",
+    "preview": "vite preview"
+  },
+  "dependencies": {
+    "@tailwindcss/vite": "^4.1.8",
+    "@tanstack/react-query": "^5.80.6",
+    "react": "^19.1.0",
+    "react-dom": "^19.1.0",
+    "react-router-dom": "^7.6.2",
+    "tailwindcss": "^4.1.8"
+  },
+  "devDependencies": {
+    "@eslint/js": "^9.25.0",
+    "@types/react": "^19.1.2",
+    "@types/react-dom": "^19.1.2",
+    "@vitejs/plugin-react": "^4.4.1",
+    "eslint": "^9.25.0",
+    "eslint-plugin-react-hooks": "^5.2.0",
+    "eslint-plugin-react-refresh": "^0.4.19",
+    "globals": "^16.0.0",
+    "typescript": "~5.8.3",
+    "typescript-eslint": "^8.30.1",
+    "vite": "^6.3.5"
+  }
+}

ui/public/site.webmanifest

@@ -0,0 +1,21 @@
+{
+  "name": "llama-swap",
+  "short_name": "llama-swap",
+  "icons": [
+    {
+      "src": "/web-app-manifest-192x192.png",
+      "sizes": "192x192",
+      "type": "image/png",
+      "purpose": "maskable"
+    },
+    {
+      "src": "/web-app-manifest-512x512.png",
+      "sizes": "512x512",
+      "type": "image/png",
+      "purpose": "maskable"
+    }
+  ],
+  "theme_color": "#ffffff",
+  "background_color": "#ffffff",
+  "display": "standalone"
+}

ui/src/App.css

@@ -0,0 +1,6 @@
+#root {
+  max-width: 1280px;
+  margin: 0 auto;
+  padding: 2rem;
+  text-align: center;
+}

ui/src/App.tsx

@@ -0,0 +1,44 @@
+import { BrowserRouter as Router, Routes, Route, Navigate, NavLink } from "react-router-dom";
+import { useTheme } from "./contexts/ThemeProvider";
+import { APIProvider } from "./contexts/APIProvider";
+import LogViewerPage from "./pages/LogViewer";
+import ModelPage from "./pages/Models";
+
+function App() {
+  const theme = useTheme();
+  return (
+    <Router basename="/ui/">
+      <APIProvider>
+        <div>
+          <nav className="bg-surface border-b border-border p-2 h-[75px]">
+            <div className="flex items-center justify-between mx-auto px-4 h-full">
+              <h1 className="flex items-center p-0">llama-swap</h1>
+              <div className="flex items-center space-x-4">
+                <NavLink to="/" className={({ isActive }) => (isActive ? "navlink active" : "navlink")}>
+                  Logs
+                </NavLink>
+
+                <NavLink to="/models" className={({ isActive }) => (isActive ? "navlink active" : "navlink")}>
+                  Models
+                </NavLink>
+                <button className="btn btn--sm" onClick={theme.toggleTheme}>
+                  {theme.isDarkMode ? "🌙" : "☀️"}
+                </button>
+              </div>
+            </div>
+          </nav>
+
+          <main className="mx-auto py-4 px-4">
+            <Routes>
+              <Route path="/" element={<LogViewerPage />} />
+              <Route path="/models" element={<ModelPage />} />
+              <Route path="*" element={<Navigate to="/" replace />} />
+            </Routes>
+          </main>
+        </div>
+      </APIProvider>
+    </Router>
+  );
+}
+
+export default App;

ui/src/assets/react.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" class="iconify iconify--logos" width="35.93" height="32" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 228"><path fill="#00D8FF" d="M210.483 73.824a171.49 171.49 0 0 0-8.24-2.597c.465-1.9.893-3.777 1.273-5.621c6.238-30.281 2.16-54.676-11.769-62.708c-13.355-7.7-35.196.329-57.254 19.526a171.23 171.23 0 0 0-6.375 5.848a155.866 155.866 0 0 0-4.241-3.917C100.759 3.829 77.587-4.822 63.673 3.233C50.33 10.957 46.379 33.89 51.995 62.588a170.974 170.974 0 0 0 1.892 8.48c-3.28.932-6.445 1.924-9.474 2.98C17.309 83.498 0 98.307 0 113.668c0 15.865 18.582 31.778 46.812 41.427a145.52 145.52 0 0 0 6.921 2.165a167.467 167.467 0 0 0-2.01 9.138c-5.354 28.2-1.173 50.591 12.134 58.266c13.744 7.926 36.812-.22 59.273-19.855a145.567 145.567 0 0 0 5.342-4.923a168.064 168.064 0 0 0 6.92 6.314c21.758 18.722 43.246 26.282 56.54 18.586c13.731-7.949 18.194-32.003 12.4-61.268a145.016 145.016 0 0 0-1.535-6.842c1.62-.48 3.21-.974 4.76-1.488c29.348-9.723 48.443-25.443 48.443-41.52c0-15.417-17.868-30.326-45.517-39.844Zm-6.365 70.984c-1.4.463-2.836.91-4.3 1.345c-3.24-10.257-7.612-21.163-12.963-32.432c5.106-11 9.31-21.767 12.459-31.957c2.619.758 5.16 1.557 7.61 2.4c23.69 8.156 38.14 20.213 38.14 29.504c0 9.896-15.606 22.743-40.946 31.14Zm-10.514 20.834c2.562 12.94 2.927 24.64 1.23 33.787c-1.524 8.219-4.59 13.698-8.382 15.893c-8.067 4.67-25.32-1.4-43.927-17.412a156.726 156.726 0 0 1-6.437-5.87c7.214-7.889 14.423-17.06 21.459-27.246c12.376-1.098 24.068-2.894 34.671-5.345a134.17 134.17 0 0 1 1.386 6.193ZM87.276 214.515c-7.882 2.783-14.16 2.863-17.955.675c-8.075-4.657-11.432-22.636-6.853-46.752a156.923 156.923 0 0 1 1.869-8.499c10.486 2.32 22.093 3.988 34.498 4.994c7.084 9.967 14.501 19.128 21.976 27.15a134.668 134.668 0 0 1-4.877 4.492c-9.933 8.682-19.886 14.842-28.658 17.94ZM50.35 144.747c-12.483-4.267-22.792-9.812-29.858-15.863c-6.35-5.437-9.555-10.836-9.555-15.216c0-9.322 13.897-21.212 37.076-29.293c2.813-.98 5.757-1.905 8.812-2.773c3.204 10.42 7.406 21.315 12.477 32.332c-5.137 11.18-9.399 22.249-12.634 32.792a134.718 134.718 0 0 1-6.318-1.979Zm12.378-84.26c-4.811-24.587-1.616-43.134 6.425-47.789c8.564-4.958 27.502 2.111 47.463 19.835a144.318 144.318 0 0 1 3.841 3.545c-7.438 7.987-14.787 17.08-21.808 26.988c-12.04 1.116-23.565 2.908-34.161 5.309a160.342 160.342 0 0 1-1.76-7.887Zm110.427 27.268a347.8 347.8 0 0 0-7.785-12.803c8.168 1.033 15.994 2.404 23.343 4.08c-2.206 7.072-4.956 14.465-8.193 22.045a381.151 381.151 0 0 0-7.365-13.322Zm-45.032-43.861c5.044 5.465 10.096 11.566 15.065 18.186a322.04 322.04 0 0 0-30.257-.006c4.974-6.559 10.069-12.652 15.192-18.18ZM82.802 87.83a323.167 323.167 0 0 0-7.227 13.238c-3.184-7.553-5.909-14.98-8.134-22.152c7.304-1.634 15.093-2.97 23.209-3.984a321.524 321.524 0 0 0-7.848 12.897Zm8.081 65.352c-8.385-.936-16.291-2.203-23.593-3.793c2.26-7.3 5.045-14.885 8.298-22.6a321.187 321.187 0 0 0 7.257 13.246c2.594 4.48 5.28 8.868 8.038 13.147Zm37.542 31.03c-5.184-5.592-10.354-11.779-15.403-18.433c4.902.192 9.899.29 14.978.29c5.218 0 10.376-.117 15.453-.343c-4.985 6.774-10.018 12.97-15.028 18.486Zm52.198-57.817c3.422 7.8 6.306 15.345 8.596 22.52c-7.422 1.694-15.436 3.058-23.88 4.071a382.417 382.417 0 0 0 7.859-13.026a347.403 347.403 0 0 0 7.425-13.565Zm-16.898 8.101a358.557 358.557 0 0 1-12.281 19.815a329.4 329.4 0 0 1-23.444.823c-7.967 0-15.716-.248-23.178-.732a310.202 310.202 0 0 1-12.513-19.846h.001a307.41 307.41 0 0 1-10.923-20.627a310.278 310.278 0 0 1 10.89-20.637l-.001.001a307.318 307.318 0 0 1 12.413-19.761c7.613-.576 15.42-.876 23.31-.876H128c7.926 0 15.743.303 23.354.883a329.357 329.357 0 0 1 12.335 19.695a358.489 358.489 0 0 1 11.036 20.54a329.472 329.472 0 0 1-11 20.722Zm22.56-122.124c8.572 4.944 11.906 24.881 6.52 51.026c-.344 1.668-.73 3.367-1.15 5.09c-10.622-2.452-22.155-4.275-34.23-5.408c-7.034-10.017-14.323-19.124-21.64-27.008a160.789 160.789 0 0 1 5.888-5.4c18.9-16.447 36.564-22.941 44.612-18.3ZM128 90.808c12.625 0 22.86 10.235 22.86 22.86s-10.235 22.86-22.86 22.86s-22.86-10.235-22.86-22.86s10.235-22.86 22.86-22.86Z"></path></svg>

ui/src/contexts/APIProvider.tsx

@@ -0,0 +1,227 @@
+import { useRef, createContext, useState, useContext, useEffect, useCallback, useMemo, type ReactNode } from "react";
+
+type ModelStatus = "ready" | "starting" | "stopping" | "stopped" | "shutdown" | "unknown";
+const LOG_LENGTH_LIMIT = 1024 * 100; /* 100KB of log data */
+
+export interface Model {
+  id: string;
+  state: ModelStatus;
+}
+
+interface APIProviderType {
+  models: Model[];
+  listModels: () => Promise<Model[]>;
+  unloadAllModels: () => Promise<void>;
+  loadModel: (model: string) => Promise<void>;
+  enableProxyLogs: (enabled: boolean) => void;
+  enableUpstreamLogs: (enabled: boolean) => void;
+  enableModelUpdates: (enabled: boolean) => void;
+  proxyLogs: string;
+  upstreamLogs: string;
+}
+
+const APIContext = createContext<APIProviderType | undefined>(undefined);
+type APIProviderProps = {
+  children: ReactNode;
+};
+
+export function APIProvider({ children }: APIProviderProps) {
+  const [proxyLogs, setProxyLogs] = useState("");
+  const [upstreamLogs, setUpstreamLogs] = useState("");
+  const proxyEventSource = useRef<EventSource | null>(null);
+  const upstreamEventSource = useRef<EventSource | null>(null);
+
+  const [models, setModels] = useState<Model[]>([]);
+  const modelStatusEventSource = useRef<EventSource | null>(null);
+
+  const appendLog = useCallback((newData: string, setter: React.Dispatch<React.SetStateAction<string>>) => {
+    setter((prev) => {
+      const updatedLog = prev + newData;
+      return updatedLog.length > LOG_LENGTH_LIMIT ? updatedLog.slice(-LOG_LENGTH_LIMIT) : updatedLog;
+    });
+  }, []);
+
+  const handleProxyMessage = useCallback(
+    (e: MessageEvent) => {
+      appendLog(e.data, setProxyLogs);
+    },
+    [proxyLogs, appendLog]
+  );
+
+  const handleUpstreamMessage = useCallback(
+    (e: MessageEvent) => {
+      appendLog(e.data, setUpstreamLogs);
+    },
+    [appendLog]
+  );
+
+  const enableProxyLogs = useCallback(
+    (enabled: boolean) => {
+      if (enabled) {
+        let retryCount = 0;
+        const maxRetries = 3;
+        const initialDelay = 1000; // 1 second
+
+        const connect = () => {
+          const eventSource = new EventSource("/logs/streamSSE/proxy");
+
+          eventSource.onmessage = handleProxyMessage;
+          eventSource.onerror = () => {
+            eventSource.close();
+            if (retryCount < maxRetries) {
+              retryCount++;
+              const delay = initialDelay * Math.pow(2, retryCount - 1);
+              setTimeout(connect, delay);
+            }
+          };
+
+          proxyEventSource.current = eventSource;
+        };
+
+        connect();
+      } else {
+        proxyEventSource.current?.close();
+        proxyEventSource.current = null;
+      }
+    },
+    [handleProxyMessage]
+  );
+
+  const enableUpstreamLogs = useCallback(
+    (enabled: boolean) => {
+      if (enabled) {
+        let retryCount = 0;
+        const maxRetries = 3;
+        const initialDelay = 1000; // 1 second
+
+        const connect = () => {
+          const eventSource = new EventSource("/logs/streamSSE/upstream");
+
+          eventSource.onmessage = handleUpstreamMessage;
+          eventSource.onerror = () => {
+            eventSource.close();
+            if (retryCount < maxRetries) {
+              retryCount++;
+              const delay = initialDelay * Math.pow(2, retryCount - 1);
+              setTimeout(connect, delay);
+            }
+          };
+
+          upstreamEventSource.current = eventSource;
+        };
+
+        connect();
+      } else {
+        upstreamEventSource.current?.close();
+        upstreamEventSource.current = null;
+      }
+    },
+    [handleUpstreamMessage]
+  );
+
+  const enableModelUpdates = useCallback(
+    (enabled: boolean) => {
+      if (enabled) {
+        const eventSource = new EventSource("/api/modelsSSE");
+        eventSource.onmessage = (e: MessageEvent) => {
+          try {
+            const models = JSON.parse(e.data) as Model[];
+            setModels(models);
+          } catch (e) {
+            console.error(e);
+          }
+        };
+        modelStatusEventSource.current = eventSource;
+      } else {
+        modelStatusEventSource.current?.close();
+        modelStatusEventSource.current = null;
+      }
+    },
+    [setModels]
+  );
+
+  useEffect(() => {
+    return () => {
+      proxyEventSource.current?.close();
+      upstreamEventSource.current?.close();
+      modelStatusEventSource.current?.close();
+    };
+  }, []);
+
+  const listModels = useCallback(async (): Promise<Model[]> => {
+    try {
+      const response = await fetch("/api/models/");
+      if (!response.ok) {
+        throw new Error(`HTTP error! status: ${response.status}`);
+      }
+      const data = await response.json();
+      return data || [];
+    } catch (error) {
+      console.error("Failed to fetch models:", error);
+      return []; // Return empty array as fallback
+    }
+  }, []);
+
+  const unloadAllModels = useCallback(async () => {
+    try {
+      const response = await fetch(`/api/models/unload/`, {
+        method: "POST",
+      });
+      if (!response.ok) {
+        throw new Error(`Failed to unload models: ${response.status}`);
+      }
+    } catch (error) {
+      console.error("Failed to unload models:", error);
+      throw error; // Re-throw to let calling code handle it
+    }
+  }, []);
+
+  const loadModel = useCallback(async (model: string) => {
+    try {
+      const response = await fetch(`/upstream/${model}/`, {
+        method: "GET",
+      });
+      if (!response.ok) {
+        throw new Error(`Failed to load model: ${response.status}`);
+      }
+    } catch (error) {
+      console.error("Failed to load model:", error);
+      throw error; // Re-throw to let calling code handle it
+    }
+  }, []);
+
+  const value = useMemo(
+    () => ({
+      models,
+      listModels,
+      unloadAllModels,
+      loadModel,
+      enableProxyLogs,
+      enableUpstreamLogs,
+      enableModelUpdates,
+      proxyLogs,
+      upstreamLogs,
+    }),
+    [
+      models,
+      listModels,
+      unloadAllModels,
+      loadModel,
+      enableProxyLogs,
+      enableUpstreamLogs,
+      enableModelUpdates,
+      proxyLogs,
+      upstreamLogs,
+    ]
+  );
+
+  return <APIContext.Provider value={value}>{children}</APIContext.Provider>;
+}
+
+export function useAPI() {
+  const context = useContext(APIContext);
+  if (context === undefined) {
+    throw new Error("useAPI must be used within an APIProvider");
+  }
+  return context;
+}

ui/src/contexts/ThemeProvider.tsx

@@ -0,0 +1,33 @@
+import { createContext, useContext, useEffect, type ReactNode } from "react";
+import { usePersistentState } from "../hooks/usePersistentState";
+
+type ThemeContextType = {
+  isDarkMode: boolean;
+  toggleTheme: () => void;
+};
+
+const ThemeContext = createContext<ThemeContextType | undefined>(undefined);
+
+type ThemeProviderProps = {
+  children: ReactNode;
+};
+
+export function ThemeProvider({ children }: ThemeProviderProps) {
+  const [isDarkMode, setIsDarkMode] = usePersistentState<boolean>("theme", false);
+
+  useEffect(() => {
+    document.documentElement.setAttribute("data-theme", isDarkMode ? "dark" : "light");
+  }, [isDarkMode]);
+
+  const toggleTheme = () => setIsDarkMode((prev) => !prev);
+
+  return <ThemeContext.Provider value={{ isDarkMode, toggleTheme }}>{children}</ThemeContext.Provider>;
+}
+
+export function useTheme(): ThemeContextType {
+  const context = useContext(ThemeContext);
+  if (context === undefined) {
+    throw new Error("useTheme must be used within a ThemeProvider");
+  }
+  return context;
+}

ui/src/hooks/usePersistentState.ts

@@ -0,0 +1,39 @@
+import { useState, useEffect, useCallback } from "react";
+
+export function usePersistentState<T>(key: string, initialValue: T): [T, (value: T | ((prevState: T) => T)) => void] {
+  const [state, setState] = useState<T>(() => {
+    if (typeof window === "undefined") return initialValue;
+    try {
+      const saved = localStorage.getItem(key);
+      return saved !== null ? JSON.parse(saved) : initialValue;
+    } catch (e) {
+      console.error(`Error parsing stored value for ${key}`, e);
+      return initialValue;
+    }
+  });
+
+  const setPersistentState = useCallback(
+    (value: T | ((prevState: T) => T)) => {
+      setState((prev) => {
+        const nextValue = typeof value === "function" ? (value as (prevState: T) => T)(prev) : value;
+        try {
+          localStorage.setItem(key, JSON.stringify(nextValue));
+        } catch (e) {
+          console.error(`Error saving value for ${key}`, e);
+        }
+        return nextValue;
+      });
+    },
+    [key]
+  );
+
+  useEffect(() => {
+    try {
+      localStorage.setItem(key, JSON.stringify(state));
+    } catch (e) {
+      console.error(`Error saving value for ${key}`, e);
+    }
+  }, [key, state]);
+
+  return [state, setPersistentState];
+}

ui/src/index.css

@@ -0,0 +1,168 @@
+@import "tailwindcss";
+@custom-variant dark (&:where([data-theme=dark], [data-theme=dark] *));
+
+@theme {
+  --color-background: rgba(252, 252, 249, 1);
+  --color-surface: rgba(255, 255, 253, 1);
+
+  /* text colors */
+  --color-txtmain: rgba(19, 52, 59, 1);
+  --color-txtsecondary: rgba(98, 108, 113, 1);
+  --color-navlink-active: rgba(245, 245, 245, 1);
+
+  --color-primary: rgba(50, 184, 198, 1);
+
+  --color-primary-hover: rgba(29, 116, 128, 1);
+  --color-primary-active: rgba(26, 104, 115, 1);
+  --color-secondary: rgba(94, 82, 64, 0.12);
+  --color-secondary-hover: rgba(94, 82, 64, 0.2);
+  --color-secondary-active: rgba(94, 82, 64, 0.25);
+  --color-border: rgba(94, 82, 64, 0.3);
+  --color-btn-primary-text: rgba(252, 252, 249, 1);
+  --color-card-border: rgba(94, 82, 64, 0.12);
+  --color-card-border-inner: rgba(94, 82, 64, 0.12);
+  --color-error: rgba(192, 21, 47, 1);
+  --color-success: rgba(33, 128, 141, 1);
+  --color-warning: rgb(244, 155, 0);
+  --color-info: rgba(98, 108, 113, 1);
+  --color-focus-ring: rgba(33, 128, 141, 0.4);
+  --color-select-caret: rgba(19, 52, 59, 0.8);
+  --color-btn-border: rgba(94, 82, 64, 0.7);
+}
+
+@layer theme {
+  /* over ride theme for dark mode */
+  [data-theme="dark"] {
+    --color-background: rgba(31, 33, 33, 1);
+    --color-surface: rgba(38, 40, 40, 1);
+    /* text colors */
+    --color-txtmain: rgba(245, 245, 245, 1);
+    --color-txtsecondary: rgba(167, 169, 169, 0.7);
+
+    --color-navlink-active: rgba(245, 245, 245, 1);
+
+    --color-primary: rgba(33, 128, 141, 1);
+    --color-primary-hover: rgba(45, 166, 178, 1);
+    --color-primary-active: rgba(41, 150, 161, 1);
+    --color-secondary: rgba(119, 124, 124, 0.15);
+    --color-secondary-hover: rgba(119, 124, 124, 0.25);
+    --color-secondary-active: rgba(119, 124, 124, 0.3);
+    --color-border: rgba(119, 124, 124, 0.3);
+    --color-error: rgba(255, 84, 89, 1);
+    --color-success: rgba(50, 184, 198, 1);
+    --color-warning: rgb(244, 155, 0);
+    --color-info: rgba(167, 169, 169, 1);
+    --color-focus-ring: rgba(50, 184, 198, 0.4);
+    --color-btn-primary-text: rgba(19, 52, 59, 1);
+    --color-card-border: rgba(119, 124, 124, 0.2);
+    --color-card-border-inner: rgba(119, 124, 124, 0.15);
+    --shadow-inset-sm: inset 0 1px 0 rgba(255, 255, 255, 0.1), inset 0 -1px 0 rgba(0, 0, 0, 0.15);
+    --button-border-secondary: rgba(119, 124, 124, 0.2);
+  }
+}
+
+@layer base {
+  body {
+    /* example of how colors using theme colors*/
+    @apply bg-background text-txtmain;
+  }
+
+  h1 {
+    @apply text-4xl text-txtmain font-bold pb-4;
+  }
+  h2 {
+    @apply text-3xl text-txtmain font-bold pb-4;
+  }
+  h3 {
+    @apply text-2xl text-txtmain font-bold pb-4;
+  }
+  h4 {
+    @apply text-xl text-txtmain font-bold pb-4;
+  }
+  h5 {
+    @apply text-lg text-txtmain font-bold pb-4;
+  }
+  h6 {
+    @apply text-base text-txtmain font-bold pb-4;
+  }
+}
+
+/* define CSS classes here for specific types of components */
+@layer components {
+  .container {
+    @apply px-4;
+  }
+
+  /* Navigation Header */
+
+  .navlink {
+    @apply text-txtsecondary hover:bg-secondary hover:text-txtmain rounded-lg p-2;
+  }
+  .navlink.active {
+    @apply bg-primary text-navlink-active;
+  }
+
+  /* Card component */
+  .card {
+    @apply bg-surface rounded-lg border border-card-border shadow-sm overflow-hidden p-4;
+  }
+
+  .card:hover {
+    @apply shadow-md;
+  }
+
+  .card__body {
+    @apply p-4;
+  }
+
+  .card__header,
+  .card__footer {
+    @apply p-4 border-b border-card-border-inner;
+  }
+
+  /* Status Badges */
+  .status {
+    @apply inline-block px-2 py-1 text-xs font-medium rounded-full;
+  }
+
+  .status--ready {
+    @apply bg-success/10 text-success;
+  }
+
+  .status--starting,
+  .status--stopping {
+    @apply bg-warning/10 text-warning;
+  }
+
+  .status--stopped {
+    @apply bg-error/10 text-error;
+  }
+
+  /* Buttons */
+  .btn {
+    @apply bg-surface p-2 px-4 text-sm rounded-full border border-2 transition-colors duration-200 border-btn-border;
+  }
+
+  .btn:hover {
+    cursor: pointer;
+  }
+
+  .btn--sm {
+    @apply px-2 py-0.5 text-xs;
+  }
+
+  .btn:disabled {
+    @apply opacity-50 cursor-not-allowed;
+  }
+}
+
+@layer utilities {
+  .ml-2 {
+    margin-left: 0.5rem;
+  }
+
+  .my-8 {
+    margin-top: 2rem;
+    margin-bottom: 2rem;
+  }
+}

ui/src/lib/Utils.ts

@@ -0,0 +1,18 @@
+export function processEvalTimes(text: string) {
+  const lines = text.match(/^ *eval time.*$/gm) || [];
+
+  let totalTokens = 0;
+  let totalTime = 0;
+
+  lines.forEach((line) => {
+    const tokensMatch = line.match(/\/\s*(\d+)\s*tokens/);
+    const timeMatch = line.match(/=\s*(\d+\.\d+)\s*ms/);
+
+    if (tokensMatch) totalTokens += parseFloat(tokensMatch[1]);
+    if (timeMatch) totalTime += parseFloat(timeMatch[1]);
+  });
+
+  const avgTokensPerSecond = totalTime > 0 ? totalTokens / (totalTime / 1000) : 0;
+
+  return [lines.length, totalTokens, Math.round(avgTokensPerSecond * 100) / 100];
+}

ui/src/main.tsx

@@ -0,0 +1,13 @@
+import { StrictMode } from "react";
+import { createRoot } from "react-dom/client";
+import "./index.css";
+import App from "./App.tsx";
+import { ThemeProvider } from "./contexts/ThemeProvider";
+
+createRoot(document.getElementById("root")!).render(
+  <StrictMode>
+    <ThemeProvider>
+      <App />
+    </ThemeProvider>
+  </StrictMode>
+);

ui/src/pages/LogViewer.tsx

@@ -0,0 +1,146 @@
+import { useState, useEffect, useRef, useMemo, useCallback } from "react";
+import { useAPI } from "../contexts/APIProvider";
+import { usePersistentState } from "../hooks/usePersistentState";
+
+const LogViewer = () => {
+  const { proxyLogs, upstreamLogs, enableProxyLogs, enableUpstreamLogs } = useAPI();
+
+  useEffect(() => {
+    enableProxyLogs(true);
+    enableUpstreamLogs(true);
+    return () => {
+      enableProxyLogs(false);
+      enableUpstreamLogs(false);
+    };
+  }, []);
+
+  return (
+    <div className="flex flex-col gap-5" style={{ height: "calc(100vh - 125px)" }}>
+      <LogPanel id="proxy" title="Proxy Logs" logData={proxyLogs} />
+      <LogPanel id="upstream" title="Upstream Logs" logData={upstreamLogs} />
+    </div>
+  );
+};
+
+interface LogPanelProps {
+  id: string;
+  title: string;
+  logData: string;
+  className?: string;
+}
+
+export const LogPanel = ({ id, title, logData, className }: LogPanelProps) => {
+  const [isCollapsed, setIsCollapsed] = usePersistentState(`logPanel-${id}-isCollapsed`, false);
+  const [filterRegex, setFilterRegex] = useState("");
+  const [fontSize, setFontSize] = usePersistentState<"xxs" | "xs" | "small" | "normal">(
+    `logPanel-${id}-fontSize`,
+    "normal"
+  );
+  const [wrapText, setTextWrap] = usePersistentState(`logPanel-${id}-wrapText`, false);
+
+  const textWrapClass = useMemo(() => {
+    return wrapText ? "whitespace-pre-wrap" : "whitespace-pre";
+  }, [wrapText]);
+
+  const toggleFontSize = useCallback(() => {
+    setFontSize((prev) => {
+      switch (prev) {
+        case "xxs":
+          return "xs";
+        case "xs":
+          return "small";
+        case "small":
+          return "normal";
+        case "normal":
+          return "xxs";
+      }
+    });
+  }, []);
+
+  const fontSizeClass = useMemo(() => {
+    switch (fontSize) {
+      case "xxs":
+        return "text-[0.5rem]"; // 0.5rem (8px)
+      case "xs":
+        return "text-[0.75rem]"; // 0.75rem (12px)
+      case "small":
+        return "text-[0.875rem]"; // 0.875rem (14px)
+      case "normal":
+        return "text-base"; // 1rem (16px)
+    }
+  }, [fontSize]);
+
+  const filteredLogs = useMemo(() => {
+    if (!filterRegex) return logData;
+    try {
+      const regex = new RegExp(filterRegex, "i");
+      const lines = logData.split("\n");
+      const filtered = lines.filter((line) => regex.test(line));
+      return filtered.join("\n");
+    } catch (e) {
+      return logData; // Return unfiltered if regex is invalid
+    }
+  }, [logData, filterRegex]);
+
+  // auto scroll to bottom
+  const preTagRef = useRef<HTMLPreElement>(null);
+  useEffect(() => {
+    if (!preTagRef.current) return;
+    preTagRef.current.scrollTop = preTagRef.current.scrollHeight;
+  }, [filteredLogs]);
+
+  return (
+    <div
+      className={`bg-surface border border-border rounded-lg overflow-hidden flex flex-col ${
+        !isCollapsed && "h-full"
+      } ${className || ""}`}
+    >
+      <div className="p-4 border-b border-border bg-secondary">
+        <div className="flex flex-col md:flex-row md:items-center md:justify-between gap-4">
+          {/* Title - Always full width on mobile, normal on desktop */}
+          <div className="w-full md:w-auto" onClick={() => setIsCollapsed(!isCollapsed)}>
+            <h3 className="m-0 text-lg">{title}</h3>
+          </div>
+
+          <div className="flex flex-col sm:flex-row gap-4 w-full md:w-auto">
+            {/* Sizing Buttons - Stacks vertically on mobile */}
+            <div className="flex flex-wrap gap-2">
+              <button className="btn" onClick={toggleFontSize}>
+                font: {fontSize}
+              </button>
+              <button className="btn" onClick={() => setTextWrap((prev) => !prev)}>
+                {wrapText ? "wrap" : "wrap off"}
+              </button>
+            </div>
+
+            {/* Filtering Options - Full width on mobile, normal on desktop */}
+            <div className="flex flex-1 min-w-0 gap-2">
+              <input
+                type="text"
+                className="flex-1 min-w-[120px] text-sm border p-2 rounded"
+                placeholder="Filter logs..."
+                value={filterRegex}
+                onChange={(e) => setFilterRegex(e.target.value)}
+              />
+              <button className="btn" onClick={() => setFilterRegex("")}>
+                Clear
+              </button>
+            </div>
+          </div>
+        </div>
+      </div>
+
+      {!isCollapsed && (
+        <div className="flex-1 bg-background font-mono text-sm p-3 overflow-hidden">
+          <pre
+            ref={preTagRef}
+            className={`h-full p-4 overflow-y-auto whitespace-pre min-h-0 ${textWrapClass} ${fontSizeClass}`}
+          >
+            {filteredLogs}
+          </pre>
+        </div>
+      )}
+    </div>
+  );
+};
+export default LogViewer;

ui/src/pages/Models.tsx

@@ -0,0 +1,110 @@
+import { useState, useEffect, useCallback, useMemo } from "react";
+import { useAPI } from "../contexts/APIProvider";
+import { LogPanel } from "./LogViewer";
+import { processEvalTimes } from "../lib/Utils";
+
+export default function ModelsPage() {
+  const { models, enableModelUpdates, unloadAllModels, loadModel, upstreamLogs, enableUpstreamLogs } = useAPI();
+  const [isUnloading, setIsUnloading] = useState(false);
+
+  useEffect(() => {
+    enableModelUpdates(true);
+    enableUpstreamLogs(true);
+    return () => {
+      enableModelUpdates(false);
+      enableUpstreamLogs(false);
+    };
+  }, []);
+
+  const handleUnloadAllModels = useCallback(async () => {
+    setIsUnloading(true);
+    try {
+      await unloadAllModels();
+    } catch (e) {
+      console.error(e);
+    } finally {
+      // at least give it a second to show the unloading message
+      setTimeout(() => {
+        setIsUnloading(false);
+      }, 1000);
+    }
+  }, []);
+
+  const [totalLines, totalTokens, avgTokensPerSecond] = useMemo(() => {
+    return processEvalTimes(upstreamLogs);
+  }, [upstreamLogs]);
+
+  return (
+    <div>
+      <div className="flex flex-col md:flex-row gap-4">
+        {/* Left Column */}
+        <div className="w-full md:w-1/2 flex items-top">
+          <div className="card w-full">
+            <h2 className="">Models</h2>
+            <button className="btn" onClick={handleUnloadAllModels} disabled={isUnloading}>
+              {isUnloading ? "Unloading..." : "Unload All Models"}
+            </button>
+            <table className="w-full mt-4">
+              <thead>
+                <tr className="border-b border-primary">
+                  <th className="text-left p-2">Name</th>
+                  <th className="text-left p-2"></th>
+                  <th className="text-left p-2">State</th>
+                </tr>
+              </thead>
+              <tbody>
+                {models.map((model) => (
+                  <tr key={model.id} className="border-b hover:bg-secondary-hover border-border">
+                    <td className="p-2">
+                      <a href={`/upstream/${model.id}/`} className="underline" target="_blank">
+                        {model.id}
+                      </a>
+                    </td>
+                    <td className="p-2">
+                      <button
+                        className="btn btn--sm"
+                        disabled={model.state !== "stopped"}
+                        onClick={() => loadModel(model.id)}
+                      >
+                        Load
+                      </button>
+                    </td>
+                    <td className="p-2">
+                      <span className={`status status--${model.state}`}>{model.state}</span>
+                    </td>
+                  </tr>
+                ))}
+              </tbody>
+            </table>
+          </div>
+        </div>
+
+        {/* Right Column */}
+        <div className="w-full md:w-1/2 flex flex-col" style={{ height: "calc(100vh - 125px)" }}>
+          <div className="card mb-4 min-h-[250px]">
+            <h2>Log Stats</h2>
+            <p className="italic my-2">note: eval logs from llama-server</p>
+            <table className="w-full border border-gray-200">
+              <tbody>
+                <tr className="border-b border-gray-200">
+                  <td className="py-2 px-4 font-medium border-r border-gray-200">Requests</td>
+                  <td className="py-2 px-4 text-right">{totalLines}</td>
+                </tr>
+                <tr className="border-b border-gray-200">
+                  <td className="py-2 px-4 font-medium border-r border-gray-200">Total Tokens Generated</td>
+                  <td className="py-2 px-4 text-right">{totalTokens}</td>
+                </tr>
+                <tr>
+                  <td className="py-2 px-4 font-medium border-r border-gray-200">Average Tokens/Second</td>
+                  <td className="py-2 px-4 text-right">{avgTokensPerSecond}</td>
+                </tr>
+              </tbody>
+            </table>
+          </div>
+
+          <LogPanel id="modelsupstream" title="Upstream Logs" logData={upstreamLogs} />
+        </div>
+      </div>
+    </div>
+  );
+}

ui/src/vite-env.d.ts

@@ -0,0 +1 @@
+/// <reference types="vite/client" />

ui/tsconfig.app.json

@@ -0,0 +1,27 @@
+{
+  "compilerOptions": {
+    "tsBuildInfoFile": "./node_modules/.tmp/tsconfig.app.tsbuildinfo",
+    "target": "ES2020",
+    "useDefineForClassFields": true,
+    "lib": ["ES2020", "DOM", "DOM.Iterable"],
+    "module": "ESNext",
+    "skipLibCheck": true,
+
+    /* Bundler mode */
+    "moduleResolution": "bundler",
+    "allowImportingTsExtensions": true,
+    "verbatimModuleSyntax": true,
+    "moduleDetection": "force",
+    "noEmit": true,
+    "jsx": "react-jsx",
+
+    /* Linting */
+    "strict": true,
+    "noUnusedLocals": true,
+    "noUnusedParameters": true,
+    "erasableSyntaxOnly": true,
+    "noFallthroughCasesInSwitch": true,
+    "noUncheckedSideEffectImports": true
+  },
+  "include": ["src"]
+}

ui/tsconfig.json

@@ -0,0 +1,7 @@
+{
+  "files": [],
+  "references": [
+    { "path": "./tsconfig.app.json" },
+    { "path": "./tsconfig.node.json" }
+  ]
+}

ui/tsconfig.node.json

@@ -0,0 +1,25 @@
+{
+  "compilerOptions": {
+    "tsBuildInfoFile": "./node_modules/.tmp/tsconfig.node.tsbuildinfo",
+    "target": "ES2022",
+    "lib": ["ES2023"],
+    "module": "ESNext",
+    "skipLibCheck": true,
+
+    /* Bundler mode */
+    "moduleResolution": "bundler",
+    "allowImportingTsExtensions": true,
+    "verbatimModuleSyntax": true,
+    "moduleDetection": "force",
+    "noEmit": true,
+
+    /* Linting */
+    "strict": true,
+    "noUnusedLocals": true,
+    "noUnusedParameters": true,
+    "erasableSyntaxOnly": true,
+    "noFallthroughCasesInSwitch": true,
+    "noUncheckedSideEffectImports": true
+  },
+  "include": ["vite.config.ts"]
+}

ui/vite.config.ts

@@ -0,0 +1,20 @@
+import { defineConfig } from "vite";
+import react from "@vitejs/plugin-react";
+import tailwindcss from "@tailwindcss/vite";
+
+// https://vite.dev/config/
+export default defineConfig({
+  plugins: [react(), tailwindcss()],
+  base: "/ui/",
+  build: {
+    outDir: "../proxy/ui_dist",
+    assetsDir: "assets",
+  },
+  server: {
+    proxy: {
+      "/api": "http://localhost:8080", // Proxy API calls to Go backend during development
+      "/logs": "http://localhost:8080",
+      "/upstream": "http://localhost:8080",
+    },
+  },
+});