vllm-omni/requirements/common.txt

# Common dependencies for all platforms
# transformers: required unified multimodal model classes are unavailable
# before 5.10.1; 5.10.0 is yanked, and 5.15 has a known construction
# regression. transformers 5.9 also changed
# _adjust_missing_and_unexpected_keys from
# list+list to set|set, breaking models that declare
# _keys_to_ignore_on_load_unexpected as a list (e.g. OpenMOSS-Team/MOSS-TTS-Nano)
# with "TypeError: unsupported operand type(s) for |: 'list' and 'set'".
# That breakage is now handled in code instead of by pinning:
# MossTTSNanoForGeneration wraps weight loading in
# transformers_keys_to_ignore_compat (vllm_omni/model_executor/models/utils.py).
# Floor is 5.13.0 so transformers.integrations.hub_kernels LayerRepository
# entries always pass version=/revision= required by kernels>=0.16 (#6971).
transformers >= 5.13.0, < 5.15
av>=14.0.0
omegaconf>=2.3.0
diffusers==0.40.0
# Hub kernels, including LTX-2.5 NATTEN. 0.16+ resolves torch-stable-abi*
# FA3 builds on torch 2.13 / cu130; keep in sync with transformers above.
kernels==0.16.1
safetensors>=0.8.0
accelerate==1.12.0
soundfile>=0.13.1
cache-dit==1.5.0
tqdm>=4.66.0
torchsde>=0.2.6
openai-whisper>=20250625
imageio[ffmpeg]>=2.37.2
x-transformers>=2.12.2
einops>=0.8.1
prettytable>=3.8.0
aenum==3.1.16
pyzmq>=25.0.0
janus>=1.0.0
msgpack>=1.0.0
pydantic>=2.1.0
pydantic-settings>=2.0.0
# Required for GGUF-quantized diffusion model loading (gguf_adapters). The CI
# base image does not ship it, so declare it explicitly. The loader also guards
# this import so a missing gguf no longer breaks all diffusion model loading.
gguf>=0.10.0
# Required for running the Cosmos3 model with guardrails enabled (default behavior).
cosmos-guardrail>=0.3.1