Files
vllm-omni/docs/features/README.md
T
2025-11-28 11:03:52 +08:00

127 B

Features

vLLM-Omni introduces new features to support multimodal comprehension and generation models inference and serving.