[WIP] Support Multimodal Input for Qwen2.5 Omni#57

Closed

tzhouam wants to merge 25 commits into

vllm-project:mainfrom

tzhouam:feat/Qwen2_5_Omni_Multimodal_Input

Commits on Nov 7, 2025

Port changes from feat/multi-request-stream (vs main)
tzhouam
committed
debug ar model runner
tzhouam
committed
Enhance error handling and logging in pipeline and stage processing. Added detailed logging for prompt loading failures, file writing issues, and stage shutdown processes. Introduced a new logging …
tzhouam
committed

Commits on Nov 10, 2025

Commits on Nov 11, 2025

Commits on Nov 12, 2025

Enhance video input handling by adding audio support in prompts and updating stage configuration to remove unnecessary process flags. This simplifies the YAML configuration for the Qwen2.5 Omni model.
tzhouam
committed