This improves the support for activity logging from the v1/responses and
v1/messages endpoints.
- add chat endpoint selection to Playground > Chat > Settings
- improve metrics extraction for streaming v1/messages and v1/responses
endpoints (tested with llama-server)
Fixes#742
- update README.md with new docker instructions
- update docs/configuration.md
- update .github/workflows to have pinned action versions
- gofmt events package
- fix small bugs in CI scripts
- reduce config options for internal/perf/monitor and config. A ring buffer is used to keep 1hr of entries at max 5s granularity. For long term stats use prometheus monitoring on /metrics
Fixes#744
## Problem
The `/running` endpoint in `listRunningProcessesHandler` reads
`process.state` directly without holding `stateMutex`. Meanwhile,
`swapState()` writes to `process.state` while holding the write lock.
This is a data race flagged by the Go race detector.
Also fixes a minor typo: "processes was in state" → "process was in
state".
## Fix
- `proxymanager.go`: Replace `process.state` with
`process.CurrentState()` which acquires `stateMutex.RLock()` before
reading.
- `process.go`: Fix typo in error message.
## Verification
- `gofmt -l` — clean
- `go test -run "TestProcessGroup_|TestProxyManager_" ./proxy/` — all
pass
- `go test ./proxy/config/... ./proxy/cache/...
./proxy/configwatcher/...` — all pass
Add a comprehensive performance monitoring system that collects CPU, memory, swap, load average, network IO, and GPU stats. Provides both a REST API for the UI and a Prometheus /metrics endpoint.
Backend changes:
- New internal/perf package with configurable interval-based stats collection
- GPU monitoring via LACT (Unix socket) and nvidia-smi fallback on Linux
- Ring buffer (internal/ring) for time-series stat storage
- Prometheus /metrics endpoint with all system and GPU metrics
- Moved LogMonitor to internal/logmon package
- New PerformanceConfig for hot-reloadable monitoring settings
- REST /api/performance endpoint replacing SSE streaming
UI changes:
- New Performance page with real-time charts for CPU, memory, GPU, and network
- Reusable PerformanceChart component
- LLAMA_SWAP_URL environment variable support
- Improved capture dialog display
Other:
- Example Grafana dashboard for Prometheus metrics
- monitor-test standalone binary
- Config schema and example updates
fixes#596
- inference handles to store an activity record for all inference endpoints
- add path, status code, and content type to Activities page
- toggle on/off columns no Activities page
- add configurable capture level for inference endpoints so large binary blobs are not stored in memory
- store captures in compressed binary format
The fsnotify-based config watcher does not work reliably when the config
file is bind-mounted into a Docker container as an individual file, and
mishandles k8s ConfigMap projections (atomically swapped symlinks).
Replace it with a small os.Stat-polling watcher and add SIGHUP as an
explicit reload signal.
- new proxy/configwatcher package: 2s os.Stat poller, follows symlinks,
fires on mtime/size change and on missing -> present transitions
- SIGHUP triggers reload unconditionally (works without --watch-config)
via the same ConfigFileChangedEvent pipeline so the UI sees identical
state transitions
- watcher goroutine now exits cleanly on shutdown via a context
- drop github.com/fsnotify/fsnotify dependency
fixes#682
- matrix.go change logic to consider any proxy.Process not in
StateStopped or StateShutdown
- process.StopImmediately, and Stop() which called it had a subtle bug
where it only handled state transitions from StateReady to
StateStopping. StateStarting -> StateStopping was ignored completely.
fix: #670
The previous captures were saved uncompressed in memory. In agentic
workflows there can be many turns with each request containing the
previous context in the body with a lot of redundant data. Use zstd to
compress the request and response data before keeping a copy of memory.
Results:
- Average Percentage Saved: 73.19%
- Average Compression Factor: ~6.77:1
I pointed Opus 4.7 (high effort) at proxy.ProcessGroup to identify any
race conditions in the swapping code. It found a race condition where
there is a small window in the fast path for routing a request to a
loaded model. There is a very small window where:
- model M1 is loaded and ready for requests
- a request, R1, for M1 comes in
- a request, R2, for M2 comes in almost immediately after
- R1 acquires the lock, sees M1 is loaded (fast path), releases the lock
`[race window]` and the request is ready to be forwarded
- the race window occurs between the release of the lock and the request
being forwarded
- the lock is released so requests can be handled concurrently
- R2 comes in within the `[race window]`, acquires the lock, triggers a
model swap to M2. stopping M1
- R1 is forwarded to a model that is unloaded or in the process of
shutting down creating an error response
In deployed systems the race window is very small and doesn't happen
often. However with #635 and PR #656 I though this deserved a bit more
attention. It is not concluded that this race is the cause of #635 but
the race is likely to happen more often under sustained or high load.
AI Note: Opus 4.7 x-high effort took about an hour to write the original
patch. With the pattern discovered the fix to matrix.go was very quick.
GLM 5.1 using the previous established patterns was able to easily write
the fix for ProcessGroup.StopProcesses().
Supersedes: #656
Updates: #277, #635
Add a new swap matrix to supersede groups for running concurrent models.
The matrix uses a solver that picks the lowest cost evictions to make a
requested model available. This simple approach along with a very basic
DSL grammar can enable very complex swapping scenarios.
- add DSL parser for set expressions with & (AND), | (OR), (), +ref
- add MatrixConfig structs, validation, and topological sort for +ref
- add MatrixSolver with cost-minimizing swap decisions
- add Matrix runtime integrating solver with Process lifecycle
- integrate matrix into ProxyManager with if-branches at all endpoints
- update config.example.yaml and config-schema.json with matrix schema
- config enforces groups XOR matrix (cannot use both)
fixes#643
Keep request duration from being underreported when upstream timings
only cover part of the full request lifecycle.
- compare wall-clock and upstream timing durations
- keep token and throughput values from timings
- add regression coverage for underreported timings
fixes#602
Add configurable HTTP timeout settings to both models and peers to support installations that requires longer timeouts than the current hardcoded defaults.
Closes#618
Add proxy routes for stable-diffusion.cpp's /sdapi/v1/txt2img,
/sdapi/v1/img2img, and /sdapi/v1/loras endpoints. POST endpoints
use proxyInferenceHandler (model in JSON body), GET /loras uses
proxyGETModelHandler (model in query param).
Update the image playground with a dual-mode UI supporting both
OpenAI and SDAPI backends. In SDAPI mode, loras are fetched first
to prime the server-side cache, and all txt2img parameters are
exposed (negative prompt, steps, cfg_scale, seed, batch_size,
clip_skip, sampler, scheduler, lora selection with multipliers).
- Add 3 sdapi route registrations in proxymanager.go
- Add sdApi.ts client with generateSdImage and fetchSdLoras
- Add SDAPI types (SdApiTxt2ImgRequest, SdApiResponse, etc.)
- Add /sdapi to vite dev proxy config
- Add backend tests for sdapi routing
- Support batch image display in gallery grid
https://claude.ai/code/session_0186MGX6NXdHVBTv2KH45fqn
---------
Co-authored-by: Claude <noreply@anthropic.com>
Extend macro substitution to the name and description fields of
ModelConfig, matching the behavior already present for cmd, proxy,
checkEndpoint, and filters.
- substitute global/model macros (including MODEL_ID) in name and
description
- substitute PORT macro in name and description when allocated
- validate no unknown macros remain in name and description after
substitution
- add tests for macro substitution, MODEL_ID, and unknown macro error
Add a new configuration parameter globalTTL that all models will
inherit. The default value is 0 which matches the currently
functionality to never automatically unload a model.
The model.ttl's default has changed to -1, which means use the global
TTL value. Any model.ttl >=0 is now value with 0 meaning never unload.
This allows a model to override a globalTTL > 0 and be configured to
never unload.
Fixes#459Closes#512
Add setParamsByID filter that applies different request parameters based
on the requested model ID, enabling per-alias behaviour for a single
loaded model.
- add SetParamsByID field to Filters struct and SanitizedSetParamsByID
method
- substitute ${MODEL_ID} and other macros in setParamsByID keys and
values
- validate no unknown macros remain in keys or values after substitution
- apply setParamsByID in proxyInferenceHandler after setParams (can
override it)
- update config-schema.json with setParamsByID definition
- update UI to show aliases and make them selectable in the Playground
closes#534
Add saving request and response headers and bodies that go through
llama-swap in memory.
- captureBuffer added to configuration. Captures are enabled by default.
- 5MB of memory is allocated for req/response captures in a ring buffer.
Setting captureBuffer to 0 will disable captures.
- UI elements to view captured data added to Activity page. Includes
some
QOL features like json formatting and recombining SSE chat streams
- capture saving is done at the byte level and has minimal impact on
llama-swap performance
Fixes#464
Ref #503
Replace the legacy React UI with the new Svelte-based one. Introduce a Playground in the UI to quickly test out text, image, text to speech and speech to text models behind llama-swap.
Key Changes
New Svelte UI (ui-svelte/)
- Multi-tab Playground with Chat, Image Generation, Audio Transcription, and Speech interfaces
- Chat: message editing/regeneration, markdown rendering with LaTeX math support, image attachments, code syntax highlighting
- Image: size selector, download/fullscreen viewing
- Audio: transcription with peer support
- Speech: voice caching with manual refresh, download button
- Responsive mobile layout with collapsible navigation
- XSS fixes and accessibility improvements
Proxy Improvements
- Add gzip/brotli compression for UI static assets (proxy/ui_compress.go)
- Add GET /v1/audio/voices?model={model} endpoint for voice listing
- Add peer support for /v1/audio/transcriptions
Extend the /running endpoint to return more details about running
processes beyond just model and state.
- add cmd field to show the command being executed
- add proxy field to show the proxy URL
- add ttl (UnloadAfter) for automatic unloading configuration
- add name and description for model metadata
- update tests to verify new fields are returned correctly
fixes#471
This commit simplifies substitution of environment variables into the configuration. There was a lot of repetitive code substituting ${env.VAR_NAME} into different fields after the configuration was parsed into a config.Config. This refactor uses a string substitution of env vars into the YAML config before it is fully parsed. This eliminates a lot of logic while maintaining backwards compatibility.
* config: support environment variable macros in peer apiKeys
Add ${env.VAR_NAME} substitution for peer apiKey fields, consistent
with existing env macro support for model fields and global apiKeys.
- Add env macro substitution for peers.{name}.apiKey in LoadConfigFromReader
- Add tests for peer apiKey env substitution
- Update config.example.yaml to show env macro usage
* config: support macros in peer apiKey and filters
Extend macro substitution to peer configuration fields:
- peers.{name}.apiKey supports both global macros and env macros
- peers.{name}.filters.stripParams supports both macro types
- peers.{name}.filters.setParams supports both macro types
Also renamed validateMetadataForUnknownMacros to validateNestedForUnknownMacros
for reuse across model metadata and peer filters validation.
Add substituteEnvMacros support for apiKeys configuration field,
allowing API keys to be loaded from environment variables using
the ${env.VAR_NAME} syntax.
- Apply env macro substitution before validation
- Add tests for env macro substitution in apiKeys
* config: add environment variable macros
Add support for ${env.VAR_NAME} syntax to pull values from system
environment variables during config loading.
- env macros processed before regular macros (allows macros to reference env vars)
- works in cmd, cmdStop, proxy, checkEndpoint, filters.stripParams, metadata
- returns error if env var is not set
- add comprehensive tests
fixes#462
* docs: add env macro example to config.example.yaml
This unifies the filtering capabilities for models and peers
- stripParams: removes params in the request
- setParams: sets params in the request
fixes#453
Fixes#444 where the UI with api keys did not work. The choice to use
http basic authorization is for simple, automatic browser support. No
changes to the UI were necessary. Just use an API key as the password,
no user name is required.
This PR allows a single llama-swap to be the central proxy for models served by other inference servers. The peer servers can be another llama-swap or any API that supports the /v1/* inference endpoint.
Updates: #433, #299Closes: #296
Add configuration support for api keys that are enforced by llama-swap. Keys are stripped before sending them to upstream servers.
Updates: #433, #50 and #251
The new logToStdout option controls what is logged to stdout. The
default has been changed to just the proxy logs, which contain swap and
http request logs.
There are four supported settings: none, proxy, upstream, both. The
"both" setting is the legacy setting where everything was spewed to
stdout.
A {model_id} containing a forward slash trips up gin's path param
parsing. This updates /logs/stream to work like /upstream where the
model_id is built up in parts and searched for in the configuration.
Updates #421
Replace container/ring.Ring with a custom circularBuffer that uses a
single contiguous []byte slice. This fixes the original implementation
which created 10,240 ring elements instead of 10KB of storage.
GetHistory is now 139x faster (145μs → 1μs) and uses 117x less memory
(1.2MB → 10KB). Allocations reduced from 2 to 1 per write operation.
Create a LogMonitor per proxy.Process, replacing the usage
of a shared one. The buffer in LogMonitor is lazy allocated on the first
call to Write and freed when the Process is stopped. This reduces
unnecessary memory usage when a model is not active.
The /logs/stream/{model_id} endpoint was added to stream logs from a
specific process.
Benson Wong
Abdulazez A.
Wim Vander Schelden
Marcus
Quentin Machu
Leoy
Ron M
Brian Mendonca
Ryan Voots
Nikesh Parajuli