[Docs] how to add a new multi stage model document (#417)

Signed-off-by: Rein Yang <ruiruyang2@gmail.com>
Signed-off-by: root <root@hk01dgx028.cm.cluster>
Co-authored-by: root <root@hk01dgx028.cm.cluster>
This commit is contained in:
rein yangandroot authored and GitHub committed 2025-12-24 15:06:29 +13:00
1 parent 2d42bdcc40
commit 2a3c50147a
4 files changed
+652 -1

No files matched your search

+19 -1
View File
@@ -1,2 +1,20 @@
# Adding a New Model
Stay tuned for guidance on how to add a new model on vLLM-Omni!
This section provides comprehensive guidance on how to add a new model to vLLM-Omni.
## Documentation
- **[Adding a New Model Guide](adding_multi_stage_model.md)**: Complete step-by-step guide using Qwen3-Omni as an example
The guide covers:
- Directory structure and organization
- Implementing stage components (thinker, talker, code2wav)
- Creating the unified model class
- Model registration
- Stage configuration
- Stage input processors
- Testing strategies
## Quick Start
For a quick reference, see the [Adding a New Model Guide](adding_multi_stage_model.md) which walks through the complete implementation of Qwen3-Omni, a multi-stage omni-modality model.