andreas/llama-swap
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
302 Commits 1 Branch 0 Tags
fc3bb716df4071ea00e9e65048857971c9573783
Commit Graph

302 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Oleg Shulyakov
fc3bb716df UI styling / code improvements (#307) 
Clean up and improve UI styling

* fix: UI - dependency cleanup
* chore: UI - start script
* refactor: UI - Extract Header
* fix: UI - Header styling
* fix: UI - LogViewer styling
* fix: UI - Models styling
* fix: UI - Activity styling
* fix: UI - ConnectionStatus colors
* review: UI - table border colors
 2025-09-19 10:47:17 -07:00
Benson Wong
c36986fef6 upstream handler support for model names with forward slash (#298) 
The upstream handler would break on model IDs that contained a forward
slash. Model IDs like "aaa/bbb" called at upstream/aaa/bbb would result
in an error. This commit adds support for model IDs with a forward slash
by iteratively searching the path for a match.

Fixes: #229
 2025-09-13 13:37:03 -07:00
Artur Podsiadły
558801db1a Fix nginx proxy buffering for streaming endpoints (#295) 
* Fix nginx proxy buffering for streaming endpoints

- Add X-Accel-Buffering: no header to SSE endpoints (/api/events, /logs/stream)
- Add X-Accel-Buffering: no header to proxied text/event-stream responses
- Add nginx reverse proxy configuration section to README
- Add tests for X-Accel-Buffering header on streaming endpoints

Fixes #236

* Fix goroutine cleanup in streaming endpoints test

Add context cancellation to TestProxyManager_StreamingEndpointsReturnNoBufferingHeader
to ensure the goroutine is properly cleaned up when the test completes.
 2025-09-09 16:07:46 -07:00
Benson Wong
b21dee27c1 Fix #288 Vite hot module reloading creating multiple SSE connections (#290) 
- move SSE (EventSource) connection to module level
- manage EventSource as a singleton, closing open connection before
  reopening a new one
 2025-09-07 21:48:58 -07:00
Benson Wong
f58c8c8ec5 Support llama.cpp's cache_n in timings info (#287) 
Capture prompt cache metrics and surface them on Activities page in UI
 2025-09-06 13:58:02 -07:00
Benson Wong
954e2dee73 Remove cmdStart from README [skip ci] 
cmdStart was in the README but it doesn't exist. Fixed the typo. Oops.
 2025-09-04 11:57:28 -07:00
Benson Wong
a533aec736 small tweak to example config 2025-09-01 21:26:58 -07:00
Brett Profitt
97b17fc47d Add ${MODEL_ID} macro (#226) 
The automatic ${MODEL_ID} macro includes the name of the model and can be used in Cmd and CmdStop.
 2025-09-01 21:21:37 -07:00
Benson Wong
2457840698 Update README.md [skip ci] 2025-08-28 23:44:37 -07:00
Benson Wong
7f55494151 Update README.md [skip ci] 2025-08-28 22:47:28 -07:00
Benson Wong
831a90d3b0 Add different timeout scenarios to Process.checkHealthEndpoint #276 (#278) 
- add a TCP connection timeout of 500ms
- increase HTTP client timeout to 5000ms

In this new behaviour the upstream has 500ms to accept a tcp connection
and 5000ms to respond to the HTTP request.
 2025-08-28 22:03:14 -07:00
Yandrik
977f1856bb add /completion endpoint (#275) 
* feat: add /completion endpoint
* chore: reformat using gofmt
 2025-08-28 21:41:02 -07:00
Benson Wong
52b329f7bc Fix #277 race condition in ProcessGroup.ProxyRequest when swap=true 2025-08-28 21:38:40 -07:00
Benson Wong
57803fd3aa Support llama-server's /infill endpoint (#272) 
Add support for llama-server's /infill endpoint and metrics gathering on the Activities page.
 2025-08-27 08:36:05 -07:00
Benson Wong
c55d0cc842 Add docs for model.concurrencyLimit #263 [skip ci] 2025-08-22 16:08:37 -07:00
Benson Wong
7acbaf4712 Add connection status indicator in UI (#260) 
* show connection status as icon in UI title
* make connection status event driven
 2025-08-20 13:58:24 -07:00
Benson Wong
fcc5ad135a UI: Allow editing of title (#246) 
- make <h1> title contentEditable
- title setting persists across reloads in localStorage
 2025-08-17 09:42:06 -07:00
Benson Wong
305e5a0031 improve example config [skip ci] 2025-08-17 09:19:04 -07:00
Benson Wong
04fc67354a Improve Activity event handling in the UI (#254) 
Improve Activity event handling in the UI

- fixes #252 found that the Activity page showed activity inconsistent
  with /api/metrics
- Change data structure for event metrics to array.
- Add Event stream connections status indicator
 2025-08-15 21:44:08 -07:00
Benson Wong
4662cf7699 add 'unconfirmed bug' as default label in bug-report.md 2025-08-15 15:38:12 -07:00
Benson Wong
5dc6b3e6d9 Add barebones but working implementation of model preload (#209, #235) 
Add barebones but working implementation of model preload

* add config test for Preload hook
* improve TestProxyManager_StartupHooks
* docs for new hook configuration
* add a .dev to .gitignore
 2025-08-14 10:27:28 -07:00
Benson Wong
74c69f39ef Add prompt processing metrics (#250) 
- capture prompt processing metrics
- display prompt processing metrics on UI Activity page
 2025-08-14 10:02:16 -07:00
Benson Wong
a186318892 Update Readme, Add screenshot for Activities page [skip ci] 2025-08-08 13:39:46 -07:00
Benson Wong
c4e4d5e1e9 Update Readme UI Screenshot [skip ci] 2025-08-08 13:33:47 -07:00
Benson Wong
7985e94ba4 add tokens processed to ui models page 2025-08-08 13:28:39 -07:00
Benson Wong
74556c3a36 Update bug-report.md [skip ci] 2025-08-08 09:52:05 -07:00
Benson Wong
5c381e4b30 Add gofmt linting to ci 2025-08-07 20:29:18 -07:00
Benson Wong
10569ed546 Fix model alias usage in upstream path (#230) 
Model alias values are not properly resolved and work in upstream/ path.

Related to #229.
 2025-08-07 20:16:56 -07:00
Benson Wong
5b10b3c23f UI Tweaks (#228) 
* sort model names in UI

* add toggle to show model id/name on UI model page
 2025-08-07 11:07:03 -07:00
Benson Wong
45ea792a3a Fix UI panel not saving position correctly 2025-08-06 14:02:22 -07:00
Benson Wong
1bc2802353 fix panels not saving sizing state 2025-08-06 14:00:21 -07:00
Benson Wong
701476c0c4 Update README.md - remove contributor block [skip ci] 
Contributor information available on the Github page's sidebar. Redundant.
 2025-08-06 11:11:47 -07:00
Ben Greene
5c63e0066c return models sorted by id in /v1/models (#222) 2025-08-06 10:04:52 -07:00
Martin Garton
8be5073c51 Fix typo (#223) [skip ci] 
Fix typo `lama-swap` -> `llama-swap`
 2025-08-06 10:02:38 -07:00
Aaron Ang
6307bd3205 Add support for building Linux ARM64 binary in Makefile (#221) 2025-08-05 16:26:06 -07:00
Benson Wong
558a72de17 UI Improvements (#219) 
- use react-resizable-panels for UI
- improve icons for buttons
- improve mobile layout with drag/resize panels
 2025-08-03 17:49:13 -07:00
Leoyzen
dc42cf366d Add config monitor support for k8s configmap. (#217) 2025-08-03 08:05:48 -07:00
Ryein Goddard
ba0a81937a Update README.md (#216) 
Update git clone protocol to https
 2025-08-01 19:48:09 -07:00
Benson Wong
574fdfabb4 UI improvements (#213) 
* use two column for logs view on wider screens

* hide log controls when panel is minimized
 2025-07-31 11:59:21 -07:00
Benson Wong
5172cb2e12 Update docs in Readme [skip ci] 2025-07-30 11:51:14 -07:00
Benson Wong
5672cb03fd Update github actions for notifying homebrew build (#212) 
Combine homebrew-llama-swap event with the release action
 2025-07-30 11:29:03 -07:00
Benson Wong
0f583163f7 add /health (#211) 2025-07-30 10:37:10 -07:00
Benson Wong
7905fa9ea3 Update trigger-homebrew-update.yml [skip ci] 2025-07-30 10:13:49 -07:00
Ian Sebastian Mathew
bbaf172956 add trigger to rebuild homebrew formula (#210) 2025-07-30 10:12:21 -07:00
Benson Wong
fd50932dbc Decouple MetricsMiddleware from downstream handlers (#206) 
* Decouple MetricsMiddleware from downstream handlers

Remove ls-real-model-name optimization. Within proxyOAIHandler the
request body's bytes are required for various rewriting features
anyways. This negated any benefits from trying not to parse it twice.
 2025-07-27 10:36:06 -07:00
Gaël James
8c693e7fcf Add endpoint aliases for reranking models (#201) 
* Add endpoint aliases for reranking models
* Add MetricsMiddleware to the previous reranking endpoint
* Fix the embeddings endpoint not having model set
 2025-07-24 08:32:47 -07:00
Benson Wong
8f2af26a41 fix stats on model page 2025-07-23 13:57:33 -07:00
Benson Wong
01d4838fb3 Fix token metrics parsing (#199) 
Fix #198

- use llama-server's `timings` info if available in response body
- send "-1" for token/sec when not able to accurately calculate
  performance
- optimize streaming body search for metrics information
 2025-07-22 23:10:14 -07:00
Benson Wong
accd65294b add contributors to README [skip ci] 2025-07-21 23:16:48 -07:00
Benson Wong
7472a25864 Update README.md [skip ci] 
update screenshot for web UI
 2025-07-21 23:08:19 -07:00