requirements.txt

# 基础框架
torch==2.2.0
torchvision==0.17.0  # 与 torch 2.4.0 兼容的版本
torchaudio==2.2.0

# 多模态处理
transformers==4.41.0
tokenizers==0.19.1  # 确保与 transformers 4.41.0 兼容
datasets==2.12.0
decord==0.6.0
librosa==0.10.2
opencv-python-headless==4.10.0.84

# 模型工具
einops==0.8.0
sentencepiece==0.2.0
accelerate==0.30.1

# 部署相关
gradio==5.12.0
fastapi==0.115.6
uvicorn==0.30.0

# 辅助工具
numpy==1.26.4
tqdm==4.66.4
requests==2.32.3
huggingface-hub==0.28.1

openai-whisper
pydantic>=1.10.0
deepspeed==0.12.2
tiktoken==0.6.0 
resampy