[build-system] requires = ["setuptools>=69.5.1"] build-backend = "setuptools.build_meta" [project] name = "llama_omni2" version = "1.0.0" description = "Towards GPT-4o like large language and speech assistant." readme = "README.md" requires-python = ">=3.10" classifiers = [ "Programming Language :: Python :: 3", "License :: OSI Approved :: Apache Software License", ] dependencies = [ "torch==2.3.1", "torchaudio==2.3.1", "transformers==4.43.4", "tokenizers==0.19.1", "sentencepiece==0.1.99", "shortuuid==1.0.13", "accelerate==0.33.0", "peft==0.14.0", "bitsandbytes==0.45.0", "pydantic==2.7.0", "markdown2[all]==2.5.2", "numpy==1.26.4", "scikit-learn==1.2.2", "gradio==5.3.0", "gradio_client==1.4.2", "requests==2.32.3", "httpx==0.28.1", "uvicorn==0.30.0", "fastapi==0.115.11", "einops==0.6.1", "einops-exts==0.0.4", "timm==0.6.13", "openai-whisper==20231117", "datasets==2.18.0", "conformer==0.3.2", "diffusers==0.27.2", "lightning==2.2.4", "gdown==5.1.0", "wget==3.2", "inflect==7.3.1", "WeTextProcessing==1.0.3", "HyperPyYAML==1.2.2", "onnxruntime-gpu==1.18.0", "omegaconf==2.3.0", "huggingface-hub==0.25.1", "hydra-core==1.3.2", "librosa==0.10.2", "onnx==1.16.0", "matplotlib==3.10.1" ] [project.optional-dependencies] train = [ "deepspeed==0.14.2", "ninja==1.11.1.3", "wandb==0.19.1", "tensorboardX==2.6.2.2" ] build = ["build", "twine"] [tool.setuptools.packages.find] include = ["llama_omni2"] [tool.wheel] universal = false