increase health check to a minimum of 5 seconds

.github/workflows/closeinactive.yml

@@ -1,23 +0,0 @@
-# https://docs.github.com/en/actions/use-cases-and-examples/project-management/closing-inactive-issues
-name: Close inactive issues
-on:
-  schedule:
-    - cron: "32 1 * * *"
-
-jobs:
-  close-issues:
-    runs-on: ubuntu-latest
-    permissions:
-      issues: write
-      pull-requests: write
-    steps:
-      - uses: actions/stale@v9
-        with:
-          days-before-issue-stale: 30
-          days-before-issue-close: 14
-          stale-issue-label: "stale"
-          stale-issue-message: "This issue is stale because it has been open for 30 days with no activity."
-          close-issue-message: "This issue was closed because it has been inactive for 14 days since being marked as stale."
-          days-before-pr-stale: -1
-          days-before-pr-close: -1
-          repo-token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/containers.yml

@@ -16,7 +16,6 @@ jobs:
     strategy:
       matrix:
         platform: [intel, cuda, vulkan, cpu, musa]
-      fail-fast: false
     steps:
       - name: Checkout code
         uses: actions/checkout@v4

.github/workflows/go-ci.yml

@@ -1,32 +0,0 @@
-# This workflow will build a golang project
-
-name: CI
-
-on:
-  push:
-    branches: [ "main" ]
-
-  pull_request:
-    branches: [ "main" ]
-
-  # Allows manual triggering of the workflow
-  workflow_dispatch:
-
-jobs:
-
-  run-tests:
-    runs-on: ubuntu-latest
-    steps:
-    - uses: actions/checkout@v4
-
-    - name: Set up Go
-      uses: actions/setup-go@v4
-      with:
-        go-version: '1.23'
-
-    # necessary for testing proxy/Process swapping
-    - name: Create simple-responder
-      run: make simple-responder
-
-    - name: Test all
-      run: make test-all

README.md

@@ -11,23 +11,20 @@ Written in golang, it is very easy to install (single binary with no dependancie
 - ✅ Easy to deploy: single binary with no dependencies
 - ✅ Easy to config: single yaml file
 - ✅ On-demand model switching
+- ✅ Full control over server settings per model
 - ✅ OpenAI API supported endpoints:
   - `v1/completions`
   - `v1/chat/completions`
   - `v1/embeddings`
   - `v1/rerank`
   - `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))
+- ✅ Multiple GPU support
-- ✅ llama-swap custom API endpoints
-  - `/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))
-  - `/running` - list currently running models ([#61](https://github.com/mostlygeek/llama-swap/issues/61))
-- ✅ Run multiple models at once with `profiles` ([docs](https://github.com/mostlygeek/llama-swap/issues/53#issuecomment-2660761741))
-- ✅ Automatic unloading of models after timeout by setting a `ttl`
-- ✅ Use any local OpenAI compatible server (llama.cpp, vllm, tabbyAPI, etc)
 - ✅ Docker and Podman support
-- ✅ Full control over server settings per model
+- ✅ Run multiple models at once with `profiles` ([docs](https://github.com/mostlygeek/llama-swap/issues/53#issuecomment-2660761741))
+- ✅ Remote log monitoring at `/log`
+- ✅ Automatic unloading of models from GPUs after timeout
+- ✅ Use any local OpenAI compatible server (llama.cpp, vllm, tabbyAPI, etc)
+- ✅ Direct access to upstream HTTP server via `/upstream/:model_id` ([demo](https://github.com/mostlygeek/llama-swap/pull/31))
 
 ## How does llama-swap work?
 
@@ -117,13 +114,6 @@ models:
       ghcr.io/ggerganov/llama.cpp:server
       --model '/models/Qwen2.5-Coder-0.5B-Instruct-Q4_K_M.gguf'
 
-  # `useModelName` will send a specific model name to the upstream server
-  # overriding whatever was set in the request
-  "qwq":
-    proxy: http://127.0.0.1:11434
-    cmd: my-server
-    useModelName: "qwen:qwq"
-
 # profiles make it easy to managing multi model (and gpu) configurations.
 #
 # Tips:
@@ -136,16 +126,11 @@ profiles:
     - "llama"
 ```
 
-### Use Case Examples
+### Advanced 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.
-
-## Configuration
-
-llama-s
 
 </details>
 
@@ -264,11 +249,3 @@ StartLimitInterval=30
 [Install]
 WantedBy=multi-user.target
 ```
-
-## Star History
-
-<picture>
-   <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/svg?repos=mostlygeek/llama-swap&type=Date&theme=dark" />
-   <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/svg?repos=mostlygeek/llama-swap&type=Date" />
-   <img alt="Star History Chart" src="https://api.star-history.com/svg?repos=mostlygeek/llama-swap&type=Date" />
-</picture>

docker/build-container.sh

@@ -38,12 +38,6 @@ else
         | jq -r --arg arch "$ARCH" '.[] | select(.metadata.container.tags[] | startswith("server-\($arch)")) | .metadata.container.tags[]' \
         | sort -r | head -n1 | awk -F '-' '{print $3}')
 
-    # Abort if LCPP_TAG is empty.
-    if [[ -z "$LCPP_TAG" ]]; then
-        echo "Abort: Could not find llama-server container for arch: $ARCH"
-        exit 1
-    fi
-
     CONTAINER_TAG="ghcr.io/mostlygeek/llama-swap:v${LS_VER}-${ARCH}-${LCPP_TAG}"
     CONTAINER_LATEST="ghcr.io/mostlygeek/llama-swap:${ARCH}"
     echo "Building ${CONTAINER_TAG} $LS_VER"

examples/restart-on-config-change/README.md

@@ -1,51 +0,0 @@
-# Restart llama-swap on config change
-
-Sometimes editing the configuration file can take a bit of trail and error to get a model configuration tuned just right. The `watch-and-restart.sh` script can be used to watch `config.yaml` for changes and restart `llama-swap` when it detects a change.
-
-```bash
-#!/bin/bash
-#
-# A simple watch and restart llama-swap when its configuration
-# file changes. Useful for trying out configuration changes
-# without manually restarting the server each time.
-if [ -z "$1" ]; then
-    echo "Usage: $0 <path to config.yaml>"
-    exit 1
-fi
-
-while true; do
-    # Start the process again
-    ./llama-swap-linux-amd64 -config $1 -listen :1867 &
-    PID=$!
-    echo "Started llama-swap with PID $PID"
-
-    # Wait for modifications in the specified directory or file
-    inotifywait -e modify "$1"
-
-    # Check if process exists before sending signal
-    if kill -0 $PID 2>/dev/null; then
-        echo "Sending SIGTERM to $PID"
-        kill -SIGTERM $PID
-        wait $PID
-    else
-        echo "Process $PID no longer exists"
-    fi
-    sleep 1
-done
-```
-
-## Usage and output example
-
-```bash
-$ ./watch-and-restart.sh config.yaml
-Started llama-swap with PID 495455
-Setting up watches.
-Watches established.
-llama-swap listening on :1867
-Sending SIGTERM to 495455
-Shutting down llama-swap
-Started llama-swap with PID 495486
-Setting up watches.
-Watches established.
-llama-swap listening on :1867
-```

go.mod

@@ -3,11 +3,7 @@ module github.com/mostlygeek/llama-swap
 go 1.23.0
 
 require (
-	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
 	gopkg.in/yaml.v3 v3.0.1
 )
 
@@ -19,10 +15,12 @@ require (
 	github.com/davecgh/go-spew v1.1.1 // indirect
 	github.com/gabriel-vasile/mimetype v1.4.3 // indirect
 	github.com/gin-contrib/sse v0.1.0 // indirect
+	github.com/gin-gonic/gin v1.10.0 // indirect
 	github.com/go-playground/locales v0.14.1 // indirect
 	github.com/go-playground/universal-translator v0.18.1 // indirect
 	github.com/go-playground/validator/v10 v10.20.0 // indirect
 	github.com/goccy/go-json v0.10.2 // indirect
+	github.com/google/shlex v0.0.0-20191202100458-e7afc7fbc510 // indirect
 	github.com/json-iterator/go v1.1.12 // indirect
 	github.com/klauspost/cpuid/v2 v2.2.7 // indirect
 	github.com/leodido/go-urn v1.4.0 // indirect
@@ -31,14 +29,12 @@ require (
 	github.com/modern-go/reflect2 v1.0.2 // indirect
 	github.com/pelletier/go-toml/v2 v2.2.2 // indirect
 	github.com/pmezard/go-difflib v1.0.0 // indirect
-	github.com/tidwall/match v1.1.1 // indirect
-	github.com/tidwall/pretty v1.2.1 // indirect
 	github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
 	github.com/ugorji/go/codec v1.2.12 // indirect
 	golang.org/x/arch v0.8.0 // indirect
-	golang.org/x/crypto v0.36.0 // indirect
+	golang.org/x/crypto v0.31.0 // indirect
-	golang.org/x/net v0.37.0 // indirect
+	golang.org/x/net v0.33.0 // indirect
-	golang.org/x/sys v0.31.0 // indirect
+	golang.org/x/sys v0.28.0 // indirect
-	golang.org/x/text v0.23.0 // indirect
+	golang.org/x/text v0.21.0 // indirect
 	google.golang.org/protobuf v1.34.1 // indirect
 )

go.sum

@@ -57,16 +57,6 @@ github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o
 github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo=
 github.com/stretchr/testify v1.9.0 h1:HtqpIVDClZ4nwg75+f6Lvsy/wHu+3BoSGCbBAcpTsTg=
 github.com/stretchr/testify v1.9.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
-github.com/tidwall/gjson v1.14.2/go.mod h1:/wbyibRr2FHMks5tjHJ5F8dMZh3AcwJEMf5vlfC0lxk=
-github.com/tidwall/gjson v1.18.0 h1:FIDeeyB800efLX89e5a8Y0BNH+LOngJyGrIWxG2FKQY=
-github.com/tidwall/gjson v1.18.0/go.mod h1:/wbyibRr2FHMks5tjHJ5F8dMZh3AcwJEMf5vlfC0lxk=
-github.com/tidwall/match v1.1.1 h1:+Ho715JplO36QYgwN9PGYNhgZvoUSc9X2c80KVTi+GA=
-github.com/tidwall/match v1.1.1/go.mod h1:eRSPERbgtNPcGhD8UCthc6PmLEQXEWd3PRB5JTxsfmM=
-github.com/tidwall/pretty v1.2.0/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
-github.com/tidwall/pretty v1.2.1 h1:qjsOFOWWQl+N3RsoF5/ssm1pHmJJwhjlSbZ51I6wMl4=
-github.com/tidwall/pretty v1.2.1/go.mod h1:ITEVvHYasfjBbM0u2Pg8T2nJnzm8xPwvNhhsoaGGjNU=
-github.com/tidwall/sjson v1.2.5 h1:kLy8mja+1c9jlljvWTlSazM7cKDRfJuR/bOJhcY5NcY=
-github.com/tidwall/sjson v1.2.5/go.mod h1:Fvgq9kS/6ociJEDnK0Fk1cpYF4FIW6ZF7LAe+6jwd28=
 github.com/twitchyliquid64/golang-asm v0.15.1 h1:SU5vSMR7hnwNxj24w34ZyCi/FmDZTkS4MhqMhdFk5YI=
 github.com/twitchyliquid64/golang-asm v0.15.1/go.mod h1:a1lVb/DtPvCB8fslRZhAngC2+aY1QWCk3Cedj/Gdt08=
 github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65EE=
@@ -78,28 +68,20 @@ golang.org/x/crypto v0.23.0 h1:dIJU/v2J8Mdglj/8rJ6UUOM3Zc9zLZxVZwwxMooUSAI=
 golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v8=
 golang.org/x/crypto v0.31.0 h1:ihbySMvVjLAeSH1IbfcRTkD/iNscyz8rGzjF/E5hV6U=
 golang.org/x/crypto v0.31.0/go.mod h1:kDsLvtWBEx7MV9tJOj9bnXsPbxwJQ6csT/x4KIN4Ssk=
-golang.org/x/crypto v0.36.0 h1:AnAEvhDddvBdpY+uR+MyHmuZzzNqXSe/GvuDeob5L34=
-golang.org/x/crypto v0.36.0/go.mod h1:Y4J0ReaxCR1IMaabaSMugxJES1EpwhBHhv2bDHklZvc=
 golang.org/x/net v0.25.0 h1:d/OCCoBEUq33pjydKrGQhw7IlUPI2Oylr+8qLx49kac=
 golang.org/x/net v0.25.0/go.mod h1:JkAGAh7GEvH74S6FOH42FLoXpXbE/aqXSrIQjXgsiwM=
 golang.org/x/net v0.33.0 h1:74SYHlV8BIgHIFC/LrYkOGIwL19eTYXQ5wc6TBuO36I=
 golang.org/x/net v0.33.0/go.mod h1:HXLR5J+9DxmrqMwG9qjGCxZ+zKXxBru04zlTvWlWuN4=
-golang.org/x/net v0.37.0 h1:1zLorHbz+LYj7MQlSf1+2tPIIgibq2eL5xkrGk6f+2c=
-golang.org/x/net v0.37.0/go.mod h1:ivrbrMbzFq5J41QOQh0siUuly180yBYtLp+CKbEaFx8=
 golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.20.0 h1:Od9JTbYCk261bKm4M/mw7AklTlFYIa0bIp9BgSm1S8Y=
 golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/sys v0.28.0 h1:Fksou7UEQUWlKvIdsqzJmUmCX3cZuD2+P3XyyzwMhlA=
 golang.org/x/sys v0.28.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
-golang.org/x/sys v0.31.0 h1:ioabZlmFYtWhL+TRYpcnNlLwhyxaM9kWTDEmfnprqik=
-golang.org/x/sys v0.31.0/go.mod h1:BJP2sWEmIv4KK5OTEluFJCKSidICx8ciO85XgH3Ak8k=
 golang.org/x/text v0.15.0 h1:h1V/4gjBv8v9cjcR6+AR5+/cIYK5N/WAgiv4xlsEtAk=
 golang.org/x/text v0.15.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
 golang.org/x/text v0.21.0 h1:zyQAAkrwaneQ066sspRyJaG9VNi/YJ1NfzcGB3hZ/qo=
 golang.org/x/text v0.21.0/go.mod h1:4IBbMaMmOPCJ8SecivzSH54+73PCFmPWxNTLm+vZkEQ=
-golang.org/x/text v0.23.0 h1:D71I7dUrlY+VX0gQShAThNGHFxZ13dGLBHQLVl1mJlY=
-golang.org/x/text v0.23.0/go.mod h1:/BLNzu4aZCJ1+kcD0DNRotWKage4q2rGVAg4o22unh4=
 google.golang.org/protobuf v1.34.1 h1:9ddQBjfCyZPOHPUiPxpYESBLc+T8P3E+Vo4IbKZgFWg=
 google.golang.org/protobuf v1.34.1/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=

misc/simple-responder/simple-responder.go

@@ -12,14 +12,12 @@ import (
 	"time"
 
 	"github.com/gin-gonic/gin"
-	"github.com/tidwall/gjson"
 )
 
 func main() {
 	gin.SetMode(gin.TestMode)
 	// Define a command-line flag for the port
 	port := flag.String("port", "8080", "port to listen on")
-	expectedModel := flag.String("model", "TheExpectedModel", "model name to expect")
 
 	// Define a command-line flag for the response message
 	responseMessage := flag.String("respond", "hi", "message to respond with")
@@ -43,70 +41,11 @@ func main() {
 		c.String(200, *responseMessage)
 	})
 
-	// for issue #62 to check model name strips profile slug
-	// has to be one of the openAI API endpoints that llama-swap proxies
-	// curl http://localhost:8080/v1/audio/speech -d '{"model":"profile:TheExpectedModel"}'
-	r.POST("/v1/audio/speech", func(c *gin.Context) {
-		body, err := io.ReadAll(c.Request.Body)
-		if err != nil {
-			c.JSON(http.StatusInternalServerError, gin.H{"error": "Failed to read request body"})
-			return
-		}
-		defer c.Request.Body.Close()
-		modelName := gjson.GetBytes(body, "model").String()
-		if modelName != *expectedModel {
-			c.JSON(http.StatusBadRequest, gin.H{"error": fmt.Sprintf("Invalid model: %s, expected: %s", modelName, *expectedModel)})
-			return
-		} else {
-			c.JSON(http.StatusOK, gin.H{"message": "ok"})
-		}
-	})
-
 	r.POST("/v1/completions", func(c *gin.Context) {
 		c.Header("Content-Type", "text/plain")
 		c.String(200, *responseMessage)
 	})
 
-	// issue #41
-	r.POST("/v1/audio/transcriptions", func(c *gin.Context) {
-		// Parse the multipart form
-		if err := c.Request.ParseMultipartForm(10 << 20); err != nil { // 10 MB max memory
-			c.JSON(http.StatusBadRequest, gin.H{"error": fmt.Sprintf("Error parsing multipart form: %s", err)})
-			return
-		}
-
-		// Get the model from the form values
-		model := c.Request.FormValue("model")
-
-		if model == "" {
-			c.JSON(http.StatusBadRequest, gin.H{"error": "Missing model parameter"})
-			return
-		}
-
-		// Get the file from the form
-		file, _, err := c.Request.FormFile("file")
-		if err != nil {
-			c.JSON(http.StatusBadRequest, gin.H{"error": fmt.Sprintf("Error getting file: %s", err)})
-			return
-		}
-		defer file.Close()
-
-		// Read the file content to get its size
-		fileBytes, err := io.ReadAll(file)
-		if err != nil {
-			c.JSON(http.StatusInternalServerError, gin.H{"error": fmt.Sprintf("Error reading file: %s", err)})
-			return
-		}
-
-		fileSize := len(fileBytes)
-
-		// Return a JSON response with the model and transcription text including file size
-		c.JSON(http.StatusOK, gin.H{
-			"text":  fmt.Sprintf("The length of the file is %d bytes", fileSize),
-			"model": model,
-		})
-	})
-
 	r.GET("/slow-respond", func(c *gin.Context) {
 		echo := c.Query("echo")
 		delay := c.Query("delay")

proxy/config.go

@@ -17,7 +17,6 @@ type ModelConfig struct {
 	CheckEndpoint string   `yaml:"checkEndpoint"`
 	UnloadAfter   int      `yaml:"ttl"`
 	Unlisted      bool     `yaml:"unlisted"`
-	UseModelName  string   `yaml:"useModelName"`
 }
 
 func (m *ModelConfig) SanitizedCommand() ([]string, error) {

proxy/process.go

@@ -116,9 +116,7 @@ func (p *Process) start() error {
 		return fmt.Errorf("can not start(), upstream proxy missing")
 	}
 
-	// multiple start() calls will wait for the one that is actually starting to
+	// wait for the other start() to complete
-	// complete before proceeding.
-	// ===========
 	curState := p.CurrentState()
 
 	if curState == StateReady {
@@ -134,26 +132,10 @@ func (p *Process) start() error {
 
 		return nil
 	}
-	// ===========
-
-	// There is the possibility of a hard to replicate race condition where
-	// curState *WAS* StateStopped but by the time we get to the p.stateMutex.Lock()
-	// below, it's value has changed!
 
 	p.stateMutex.Lock()
 	defer p.stateMutex.Unlock()
 
-	// with the exclusive lock, check if p.state is StateStopped, which is the only valid state
-	// to transition from to StateReady
-
-	if p.state != StateStopped {
-		if p.state == StateReady {
-			return nil
-		} else {
-			return fmt.Errorf("start() can not proceed expected StateReady but process is in %v", p.state)
-		}
-	}
-
 	if err := p.setState(StateStarting); err != nil {
 		return err
 	}

proxy/process_test.go

@@ -169,8 +169,6 @@ func TestProcess_LowTTLValue(t *testing.T) {
 }
 
 // issue #19
-// This test makes sure using Process.Stop() does not affect pending HTTP
-// requests. All HTTP requests in this test should complete successfully.
 func TestProcess_HTTPRequestsHaveTimeToFinish(t *testing.T) {
 	if testing.Short() {
 		t.Skip("skipping slow test")
@@ -194,9 +192,8 @@ func TestProcess_HTTPRequestsHaveTimeToFinish(t *testing.T) {
 		wg.Add(1)
 		go func(key string) {
 			defer wg.Done()
-			// send a request where simple-responder is will wait 300ms before responding
+			// send a request that should take 5 * 200ms (1 second) to complete
-			// this will simulate an in-progress request.
+			req := httptest.NewRequest("GET", fmt.Sprintf("/slow-respond?echo=%s&delay=200ms", key), nil)
-			req := httptest.NewRequest("GET", fmt.Sprintf("/slow-respond?echo=%s&delay=300ms", key), nil)
 			w := httptest.NewRecorder()
 
 			process.ProxyRequest(w, req)
@@ -212,9 +209,9 @@ func TestProcess_HTTPRequestsHaveTimeToFinish(t *testing.T) {
 		}(key)
 	}
 
-	// Stop the process while requests are still being processed
+	// stop the requests in the middle
 	go func() {
-		<-time.After(150 * time.Millisecond)
+		<-time.After(500 * time.Millisecond)
 		process.Stop()
 	}()
 

proxy/proxymanager.go

@@ -5,7 +5,6 @@ import (
 	"encoding/json"
 	"fmt"
 	"io"
-	"mime/multipart"
 	"net/http"
 	"sort"
 	"strconv"
@@ -14,8 +13,6 @@ import (
 	"time"
 
 	"github.com/gin-gonic/gin"
-	"github.com/tidwall/gjson"
-	"github.com/tidwall/sjson"
 )
 
 const (
@@ -96,7 +93,6 @@ func New(config *Config) *ProxyManager {
 
 	// Support audio/speech endpoint
 	pm.ginEngine.POST("/v1/audio/speech", pm.proxyOAIHandler)
-	pm.ginEngine.POST("/v1/audio/transcriptions", pm.proxyOAIPostFormHandler)
 
 	pm.ginEngine.GET("/v1/models", pm.listModelsHandler)
 
@@ -108,10 +104,6 @@ func New(config *Config) *ProxyManager {
 	pm.ginEngine.GET("/upstream", pm.upstreamIndex)
 	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")
@@ -230,7 +222,11 @@ func (pm *ProxyManager) swapModel(requestedModel string) (*Process, error) {
 	defer pm.Unlock()
 
 	// Check if requestedModel contains a PROFILE_SPLIT_CHAR
-	profileName, modelName := splitRequestedModel(requestedModel)
+	profileName, modelName := "", requestedModel
+	if idx := strings.Index(requestedModel, PROFILE_SPLIT_CHAR); idx != -1 {
+		profileName = requestedModel[:idx]
+		modelName = requestedModel[idx+1:]
+	}
 
 	if profileName != "" {
 		if _, found := pm.config.Profiles[profileName]; !found {
@@ -346,148 +342,29 @@ func (pm *ProxyManager) proxyOAIHandler(c *gin.Context) {
 		return
 	}
 
-	requestedModel := gjson.GetBytes(bodyBytes, "model").String()
+	var requestBody map[string]interface{}
-	if requestedModel == "" {
+	if err := json.Unmarshal(bodyBytes, &requestBody); err != nil {
+		pm.sendErrorResponse(c, http.StatusBadRequest, fmt.Sprintf("invalid JSON: %s", err.Error()))
+		return
+	}
+	model, ok := requestBody["model"].(string)
+	if !ok {
 		pm.sendErrorResponse(c, http.StatusBadRequest, "missing or invalid 'model' key")
-	}
-
-	process, err := pm.swapModel(requestedModel)
-
-	if err != nil {
-		pm.sendErrorResponse(c, http.StatusNotFound, fmt.Sprintf("unable to swap to model, %s", err.Error()))
 		return
 	}
 
-	// issue #69 allow custom model names to be sent to upstream
+	if process, err := pm.swapModel(model); err != nil {
-	if process.config.UseModelName != "" {
+		pm.sendErrorResponse(c, http.StatusNotFound, fmt.Sprintf("unable to swap to model, %s", err.Error()))
-		bodyBytes, err = sjson.SetBytes(bodyBytes, "model", process.config.UseModelName)
+		return
-		if err != nil {
-			pm.sendErrorResponse(c, http.StatusInternalServerError, fmt.Sprintf("error updating JSON: %s", err.Error()))
-			return
-		}
 	} else {
-		profileName, modelName := splitRequestedModel(requestedModel)
+		c.Request.Body = io.NopCloser(bytes.NewBuffer(bodyBytes))
-		if profileName != "" {
-			bodyBytes, err = sjson.SetBytes(bodyBytes, "model", modelName)
-			if err != nil {
-				pm.sendErrorResponse(c, http.StatusInternalServerError, fmt.Sprintf("error updating JSON: %s", err.Error()))
-				return
-			}
-		}
 
+		// 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)))
+
+		process.ProxyRequest(c.Writer, c.Request)
 	}
-
-	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)))
-
-	process.ProxyRequest(c.Writer, c.Request)
-
-}
-
-func (pm *ProxyManager) proxyOAIPostFormHandler(c *gin.Context) {
-	// We need to reconstruct the multipart form in any case since the body is consumed
-	// Create a new buffer for the reconstructed request
-	var requestBuffer bytes.Buffer
-	multipartWriter := multipart.NewWriter(&requestBuffer)
-
-	// Parse multipart form
-	if err := c.Request.ParseMultipartForm(32 << 20); err != nil { // 32MB max memory, larger files go to tmp disk
-		pm.sendErrorResponse(c, http.StatusBadRequest, fmt.Sprintf("error parsing multipart form: %s", err.Error()))
-		return
-	}
-
-	// Get model parameter from the form
-	requestedModel := c.Request.FormValue("model")
-	if requestedModel == "" {
-		pm.sendErrorResponse(c, http.StatusBadRequest, "missing or invalid 'model' parameter in form data")
-		return
-	}
-
-	// Swap to the requested model
-	process, err := pm.swapModel(requestedModel)
-	if err != nil {
-		pm.sendErrorResponse(c, http.StatusNotFound, fmt.Sprintf("unable to swap to model, %s", err.Error()))
-		return
-	}
-
-	// Get profile name and model name from the requested model
-	profileName, modelName := splitRequestedModel(requestedModel)
-
-	// Copy all form values
-	for key, values := range c.Request.MultipartForm.Value {
-		for _, value := range values {
-			fieldValue := value
-			// If this is the model field and we have a profile, use just the model name
-			if key == "model" {
-				if process.config.UseModelName != "" {
-					fieldValue = process.config.UseModelName
-				} else if profileName != "" {
-					fieldValue = modelName
-				}
-			}
-			field, err := multipartWriter.CreateFormField(key)
-			if err != nil {
-				pm.sendErrorResponse(c, http.StatusInternalServerError, "error recreating form field")
-				return
-			}
-			if _, err = field.Write([]byte(fieldValue)); err != nil {
-				pm.sendErrorResponse(c, http.StatusInternalServerError, "error writing form field")
-				return
-			}
-		}
-	}
-
-	// Copy all files from the original request
-	for key, fileHeaders := range c.Request.MultipartForm.File {
-		for _, fileHeader := range fileHeaders {
-			formFile, err := multipartWriter.CreateFormFile(key, fileHeader.Filename)
-			if err != nil {
-				pm.sendErrorResponse(c, http.StatusInternalServerError, "error recreating form file")
-				return
-			}
-
-			file, err := fileHeader.Open()
-			if err != nil {
-				pm.sendErrorResponse(c, http.StatusInternalServerError, "error opening uploaded file")
-				return
-			}
-
-			if _, err = io.Copy(formFile, file); err != nil {
-				file.Close()
-				pm.sendErrorResponse(c, http.StatusInternalServerError, "error copying file data")
-				return
-			}
-			file.Close()
-		}
-	}
-
-	// Close the multipart writer to finalize the form
-	if err := multipartWriter.Close(); err != nil {
-		pm.sendErrorResponse(c, http.StatusInternalServerError, "error finalizing multipart form")
-		return
-	}
-
-	// Create a new request with the reconstructed form data
-	modifiedReq, err := http.NewRequestWithContext(
-		c.Request.Context(),
-		c.Request.Method,
-		c.Request.URL.String(),
-		&requestBuffer,
-	)
-	if err != nil {
-		pm.sendErrorResponse(c, http.StatusInternalServerError, "error creating modified request")
-		return
-	}
-
-	// Copy the headers from the original request
-	modifiedReq.Header = c.Request.Header.Clone()
-	modifiedReq.Header.Set("Content-Type", multipartWriter.FormDataContentType())
-
-	// Use the modified request for proxying
-	process.ProxyRequest(c.Writer, modifiedReq)
 }
 
 func (pm *ProxyManager) sendErrorResponse(c *gin.Context, statusCode int, message string) {
@@ -500,42 +377,6 @@ func (pm *ProxyManager) sendErrorResponse(c *gin.Context, statusCode int, messag
 	}
 }
 
-func (pm *ProxyManager) unloadAllModelsHandler(c *gin.Context) {
-	pm.StopProcesses()
-	c.String(http.StatusOK, "OK")
-}
-
-func (pm *ProxyManager) listRunningProcessesHandler(context *gin.Context) {
-	context.Header("Content-Type", "application/json")
-	runningProcesses := make([]gin.H, 0) // Default to an empty response.
-
-	for _, process := range pm.currentProcesses {
-
-		// Append the process ID and State (multiple entries if profiles are being used).
-		runningProcesses = append(runningProcesses, gin.H{
-			"model": process.ID,
-			"state": process.state,
-		})
-
-	}
-
-	// Put the results under the `running` key.
-	response := gin.H{
-		"running": runningProcesses,
-	}
-
-	context.JSON(http.StatusOK, response) // Always return 200 OK
-}
-
 func ProcessKeyName(groupName, modelName string) string {
 	return groupName + PROFILE_SPLIT_CHAR + modelName
 }
-
-func splitRequestedModel(requestedModel string) (string, string) {
-	profileName, modelName := "", requestedModel
-	if idx := strings.Index(requestedModel, PROFILE_SPLIT_CHAR); idx != -1 {
-		profileName = requestedModel[:idx]
-		modelName = requestedModel[idx+1:]
-	}
-	return profileName, modelName
-}

proxy/proxymanager_test.go

@@ -4,8 +4,6 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"math/rand"
-	"mime/multipart"
 	"net/http"
 	"net/http/httptest"
 	"sync"
@@ -306,338 +304,3 @@ func TestProxyManager_Shutdown(t *testing.T) {
 	}()
 	wg.Wait()
 }
-
-func TestProxyManager_Unload(t *testing.T) {
-	config := &Config{
-		HealthCheckTimeout: 15,
-		Models: map[string]ModelConfig{
-			"model1": getTestSimpleResponderConfig("model1"),
-		},
-	}
-
-	proxy := New(config)
-	proc, err := proxy.swapModel("model1")
-	assert.NoError(t, err)
-	assert.NotNil(t, proc)
-
-	assert.Len(t, proxy.currentProcesses, 1)
-	req := httptest.NewRequest("GET", "/unload", nil)
-	w := httptest.NewRecorder()
-	proxy.HandlerFunc(w, req)
-	assert.Equal(t, http.StatusOK, w.Code)
-	assert.Equal(t, w.Body.String(), "OK")
-	assert.Len(t, proxy.currentProcesses, 0)
-}
-
-// issue 62, strip profile slug from model name
-func TestProxyManager_StripProfileSlug(t *testing.T) {
-	config := &Config{
-		HealthCheckTimeout: 15,
-		Profiles: map[string][]string{
-			"test": {"TheExpectedModel"}, // TheExpectedModel is default in simple-responder.go
-		},
-		Models: map[string]ModelConfig{
-			"TheExpectedModel": getTestSimpleResponderConfig("TheExpectedModel"),
-		},
-	}
-
-	proxy := New(config)
-	defer proxy.StopProcesses()
-
-	reqBody := fmt.Sprintf(`{"model":"%s"}`, "test:TheExpectedModel")
-	req := httptest.NewRequest("POST", "/v1/audio/speech", bytes.NewBufferString(reqBody))
-	w := httptest.NewRecorder()
-	proxy.HandlerFunc(w, req)
-	assert.Equal(t, http.StatusOK, w.Code)
-	assert.Contains(t, w.Body.String(), "ok")
-}
-
-// Test issue #61 `Listing the current list of models and the loaded model.`
-func TestProxyManager_RunningEndpoint(t *testing.T) {
-
-	// Shared configuration
-	config := &Config{
-		HealthCheckTimeout: 15,
-		Models: map[string]ModelConfig{
-			"model1": getTestSimpleResponderConfig("model1"),
-			"model2": getTestSimpleResponderConfig("model2"),
-		},
-		Profiles: map[string][]string{
-			"test": {"model1", "model2"},
-		},
-	}
-
-	// Define a helper struct to parse the JSON response.
-	type RunningResponse struct {
-		Running []struct {
-			Model string `json:"model"`
-			State string `json:"state"`
-		} `json:"running"`
-	}
-
-	// Create proxy once for all tests
-	proxy := New(config)
-	defer proxy.StopProcesses()
-
-	t.Run("no models loaded", func(t *testing.T) {
-		req := httptest.NewRequest("GET", "/running", nil)
-		w := httptest.NewRecorder()
-		proxy.HandlerFunc(w, req)
-
-		assert.Equal(t, http.StatusOK, w.Code)
-
-		var response RunningResponse
-
-		// Check if this is a valid JSON object.
-		assert.NoError(t, json.Unmarshal(w.Body.Bytes(), &response))
-
-		// We should have an empty running array here.
-		assert.Empty(t, response.Running, "expected no running models")
-	})
-
-	t.Run("single model loaded", func(t *testing.T) {
-		// Load just a model.
-		reqBody := `{"model":"model1"}`
-		req := httptest.NewRequest("POST", "/v1/chat/completions", bytes.NewBufferString(reqBody))
-		w := httptest.NewRecorder()
-		proxy.HandlerFunc(w, req)
-		assert.Equal(t, http.StatusOK, w.Code)
-
-		// Simulate browser call for the `/running` endpoint.
-		req = httptest.NewRequest("GET", "/running", nil)
-		w = httptest.NewRecorder()
-		proxy.HandlerFunc(w, req)
-
-		var response RunningResponse
-		assert.NoError(t, json.Unmarshal(w.Body.Bytes(), &response))
-
-		// Check if we have a single array element.
-		assert.Len(t, response.Running, 1)
-
-		// Is this the right model?
-		assert.Equal(t, "model1", response.Running[0].Model)
-
-		// Is the model loaded?
-		assert.Equal(t, "ready", response.Running[0].State)
-	})
-
-	t.Run("multiple models via profile", func(t *testing.T) {
-		// Load more than one model.
-		for _, model := range []string{"model1", "model2"} {
-			profileModel := ProcessKeyName("test", model)
-			reqBody := fmt.Sprintf(`{"model":"%s"}`, profileModel)
-			req := httptest.NewRequest("POST", "/v1/chat/completions", bytes.NewBufferString(reqBody))
-			w := httptest.NewRecorder()
-			proxy.HandlerFunc(w, req)
-			assert.Equal(t, http.StatusOK, w.Code)
-		}
-
-		// Simulate the browser call.
-		req := httptest.NewRequest("GET", "/running", nil)
-		w := httptest.NewRecorder()
-		proxy.HandlerFunc(w, req)
-
-		var response RunningResponse
-
-		// The JSON response must be valid.
-		assert.NoError(t, json.Unmarshal(w.Body.Bytes(), &response))
-
-		// The response should contain 2 models.
-		assert.Len(t, response.Running, 2)
-
-		expectedModels := map[string]struct{}{
-			"model1": {},
-			"model2": {},
-		}
-
-		// Iterate through the models and check their states as well.
-		for _, entry := range response.Running {
-			_, exists := expectedModels[entry.Model]
-			assert.True(t, exists, "unexpected model %s", entry.Model)
-			assert.Equal(t, "ready", entry.State)
-			delete(expectedModels, entry.Model)
-		}
-
-		// Since we deleted each model while testing for its validity we should have no more models in the response.
-		assert.Empty(t, expectedModels, "unexpected additional models in response")
-	})
-}
-
-func TestProxyManager_AudioTranscriptionHandler(t *testing.T) {
-	config := &Config{
-		HealthCheckTimeout: 15,
-		Profiles: map[string][]string{
-			"test": {"TheExpectedModel"},
-		},
-		Models: map[string]ModelConfig{
-			"TheExpectedModel": getTestSimpleResponderConfig("TheExpectedModel"),
-		},
-	}
-
-	proxy := New(config)
-	defer proxy.StopProcesses()
-
-	testCases := []struct {
-		name        string
-		modelInput  string
-		expectModel string
-	}{
-		{
-			name:        "With Profile Prefix",
-			modelInput:  "test:TheExpectedModel",
-			expectModel: "TheExpectedModel", // Profile prefix should be stripped
-		},
-		{
-			name:        "Without Profile Prefix",
-			modelInput:  "TheExpectedModel",
-			expectModel: "TheExpectedModel", // Should remain the same
-		},
-	}
-
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			// Create a buffer with multipart form data
-			var b bytes.Buffer
-			w := multipart.NewWriter(&b)
-
-			// Add the model field
-			fw, err := w.CreateFormField("model")
-			assert.NoError(t, err)
-			_, err = fw.Write([]byte(tc.modelInput))
-			assert.NoError(t, err)
-
-			// Add a file field
-			fw, err = w.CreateFormFile("file", "test.mp3")
-			assert.NoError(t, err)
-			// Generate random content length between 10 and 20
-			contentLength := rand.Intn(11) + 10 // 10 to 20
-			content := make([]byte, contentLength)
-			_, err = fw.Write(content)
-			assert.NoError(t, err)
-			w.Close()
-
-			// Create the request with the multipart form data
-			req := httptest.NewRequest("POST", "/v1/audio/transcriptions", &b)
-			req.Header.Set("Content-Type", w.FormDataContentType())
-			rec := httptest.NewRecorder()
-			proxy.HandlerFunc(rec, req)
-
-			// Verify the response
-			assert.Equal(t, http.StatusOK, rec.Code)
-			var response map[string]string
-			err = json.Unmarshal(rec.Body.Bytes(), &response)
-			assert.NoError(t, err)
-			assert.Equal(t, tc.expectModel, response["model"])
-			assert.Equal(t, response["text"], fmt.Sprintf("The length of the file is %d bytes", contentLength)) // matches simple-responder
-		})
-	}
-}
-
-func TestProxyManager_SplitRequestedModel(t *testing.T) {
-
-	tests := []struct {
-		name            string
-		requestedModel  string
-		expectedProfile string
-		expectedModel   string
-	}{
-		{"no profile", "gpt-4", "", "gpt-4"},
-		{"with profile", "profile1:gpt-4", "profile1", "gpt-4"},
-		{"only profile", "profile1:", "profile1", ""},
-		{"empty model", ":gpt-4", "", "gpt-4"},
-		{"empty profile", ":", "", ""},
-		{"no split char", "gpt-4", "", "gpt-4"},
-		{"profile and model with delimiter", "profile1:delimiter:gpt-4", "profile1", "delimiter:gpt-4"},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			profileName, modelName := splitRequestedModel(tt.requestedModel)
-			if profileName != tt.expectedProfile {
-				t.Errorf("splitRequestedModel(%q) = %q, %q; want %q, %q", tt.requestedModel, profileName, modelName, tt.expectedProfile, tt.expectedModel)
-			}
-			if modelName != tt.expectedModel {
-				t.Errorf("splitRequestedModel(%q) = %q, %q; want %q, %q", tt.requestedModel, profileName, modelName, tt.expectedProfile, tt.expectedModel)
-			}
-		})
-	}
-}
-
-// 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
-
-	config := &Config{
-		HealthCheckTimeout: 15,
-		Profiles: map[string][]string{
-			"test": {"model1"},
-		},
-
-		Models: map[string]ModelConfig{
-			"model1": modelConfig,
-		},
-	}
-
-	proxy := New(config)
-	defer proxy.StopProcesses()
-
-	tests := []struct {
-		description    string
-		requestedModel string
-	}{
-		{"useModelName over rides requested model", "model1"},
-		{"useModelName over rides requested profile:model", "test:model1"},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.description+": /v1/chat/completions", func(t *testing.T) {
-			reqBody := fmt.Sprintf(`{"model":"%s"}`, tt.requestedModel)
-			req := httptest.NewRequest("POST", "/v1/chat/completions", bytes.NewBufferString(reqBody))
-			w := httptest.NewRecorder()
-
-			proxy.HandlerFunc(w, req)
-			assert.Equal(t, http.StatusOK, w.Code)
-			assert.Contains(t, w.Body.String(), upstreamModelName)
-
-		})
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.description+": /v1/audio/transcriptions", func(t *testing.T) {
-			// Create a buffer with multipart form data
-			var b bytes.Buffer
-			w := multipart.NewWriter(&b)
-
-			// Add the model field
-			fw, err := w.CreateFormField("model")
-			assert.NoError(t, err)
-			_, err = fw.Write([]byte(tt.requestedModel))
-			assert.NoError(t, err)
-
-			// Add a file field
-			fw, err = w.CreateFormFile("file", "test.mp3")
-			assert.NoError(t, err)
-			_, err = fw.Write([]byte("test"))
-			assert.NoError(t, err)
-			w.Close()
-
-			// Create the request with the multipart form data
-			req := httptest.NewRequest("POST", "/v1/audio/transcriptions", &b)
-			req.Header.Set("Content-Type", w.FormDataContentType())
-			rec := httptest.NewRecorder()
-			proxy.HandlerFunc(rec, req)
-
-			// Verify the response
-			assert.Equal(t, http.StatusOK, rec.Code)
-			var response map[string]string
-			err = json.Unmarshal(rec.Body.Bytes(), &response)
-			assert.NoError(t, err)
-			assert.Equal(t, upstreamModelName, response["model"])
-		})
-	}
-
-}