Labyricorn/voicebox
1
0
Fork 0
mirror of https://github.com/jamiepine/voicebox.git synced 2026-10-03 17:15:19 -07:00
Code Issues Packages Projects Releases Wiki Activity

fix(models): honor model_name on /models/load and /models/unload (fixes #977)

Browse Source
This commit is contained in:
SEPURI-SAI-KRISHNA
2026-10-04 00:01:00 +00:00
committed by capy-ai-staging[bot]
parent d09c5c39e3
commit cf0da2e6a3
4 changed files with 218 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/content/docs/developer/model-management.mdx
+4 -2
View File
@@ -117,7 +117,9 @@ POST /models/load
}
```
The route looks up the config, dispatches to `get_model_load_func(config)`, and returns once the model is ready.
The route looks up the config, dispatches to `get_model_load_func(config)`, and returns once the model is ready. `model_name` is any id from `GET /models/status` — TTS, Whisper, and LLM entries all resolve through the same registry.
Omitting `model_name` falls back to the default Qwen TTS backend, selected by an optional `model_size` (`POST /models/load?model_size=0.6B`). This is the pre-`model_name` form and still works.
### Unload
@@ -128,7 +130,7 @@ POST /models/unload
}
```
Calls `unload_model_by_config(config)`, which routes to the right backend's `unload_model()` and frees GPU memory.
Calls `unload_model_by_config(config)`, which routes to the right backend's `unload_model()` and frees GPU memory. `POST /models/{name}/unload` is equivalent. Omitting `model_name` unloads the default Qwen TTS model.
### Download