32 Commits

Author SHA1 Message Date
219db7abce Move port range validation to config 2025-10-27 18:23:49 +01:00
72586fc627 Simplify config tests 2025-10-19 19:06:06 +02:00
6a91fe13e0 Fix local node override tests 2025-10-19 18:59:59 +02:00
82f4f7beed Ensure local node is defined in LoadConfig by adding default config if missing 2025-10-19 18:47:02 +02:00
867380a06d Remove GetBackendSettings method from config 2025-10-19 18:32:05 +02:00
688b815ca7 Add LocalNode configuration 2025-10-08 19:43:53 +02:00
7f6725da96 Refactor NodeConfig handling to use a map 2025-10-08 19:24:24 +02:00
16b28bac05 Merge branch 'main' into feat/multi-host 2025-10-07 18:04:24 +02:00
Anuruth Lertpiya
0e1bc8a352 Added support for configuring CORS headers 2025-10-04 09:13:40 +00:00
2759be65a5 Add remote instance management functionality and configuration support 2025-09-30 21:09:05 +02:00
Anuruth Lertpiya
5906d89f8d Added support for serving behind a reverse proxy
- Added support for specifying response headers for each backend
  - Allowing users to set `X-Accel-Buffering: no` to disable buffering for streaming responses in nginx
  - Updated `configuration.md` to document the new configuration options
- Modified Vite config to build with `base: "./"`, making assets be accessed via relative paths
- Updated API_BASE to use `document.baseURI`, allowing API calls to be made relative to the base path
2025-09-29 12:43:10 +00:00
fa9335663a Parse backend env vars from env vars 2025-09-28 15:22:01 +02:00
50e1355205 Add environment field to BackendSettings for improved configuration 2025-09-28 14:42:10 +02:00
Anuruth Lertpiya
f496a28f04 fix: InstancesDir and LogsDir not being relative path to DataDir when not set 2025-09-27 18:14:25 +00:00
Anuruth Lertpiya
3a979da815 fix: llamactl reads config file per documentation
- Added logging to track config file reading operations
- llamactl now properly reads config files from the expected locations ("llamactl.yaml" and "config.yaml" under current directory)
2025-09-27 17:03:54 +00:00
031d6c7017 Update Docker command arguments for llama-server and vllm with volume mounts 2025-09-25 22:51:51 +02:00
72d2a601c8 Update Docker args in LoadConfig and tests to include 'run --rm' prefix 2025-09-24 21:27:51 +02:00
9a56660f68 Refactor backend configuration to use structured settings and update environment variable handling 2025-09-24 20:31:20 +02:00
4df02a6519 Initial vLLM backend support 2025-09-19 18:05:12 +02:00
6a580667ed Remove LlamaExecutable checks from default and file loading tests 2025-09-18 20:30:26 +02:00
5121f0e302 Remove PythonPath references from MlxServerOptions and related configurations 2025-09-17 21:59:55 +02:00
cc5d8acd92 Refactor instance and manager tests to use BackendConfig for LlamaExecutable and MLXLMExecutable 2025-09-16 21:45:50 +02:00
468688cdbc Pass backend options to instances 2025-09-16 21:37:48 +02:00
988c4aca40 Add MLX backend config options 2025-09-16 21:14:19 +02:00
a9f1c1a619 Add LRU eviction configuration for instances 2025-08-30 22:26:02 +02:00
615c2ac54e Add MaxRunningInstances to InstancesConfig and implement IsRunning method 2025-08-27 18:42:34 +02:00
ddb54763f6 Add OnDemandStartTimeout configuration and update OpenAIProxy to use it 2025-08-20 14:25:43 +02:00
7b4adfa0cd Add DefaultOnDemandStart configuration and update instance options 2025-08-20 13:50:43 +02:00
ccffbca6b2 Add timeout check interval and update instance configuration 2025-08-17 19:26:21 +02:00
e2b64620b5 Expose version endpoint 2025-08-07 19:10:06 +02:00
2abe9c282e Rename config and instance struct to avoid awkward naming 2025-08-04 19:30:50 +02:00
6a7a9a2d09 Split large package into subpackages 2025-08-04 19:23:56 +02:00