Update requirements and refactor client integration: Add extra index URL for PyTorch in requirements.txt, integrate open_ai_client in main.py, and adjust image generation parameters in image_service.py. Refactor llama_cpp_client to improve model loading configuration and enhance error handling in image_pipeline_client.
32efff5
| fastapi[standard] == 0.114.0 | |
| uvicorn == 0.34.2 | |
| requests == 2.32.3 | |
| huggingface-hub == 0.32.0 | |
| # If use diffusers | |
| diffusers == 0.33.1 | |
| accelerate == 1.6.0 | |
| # transformers == 4.52.4 | |
| torch==2.7.0 | |
| --extra-index-url https://download.pytorch.org/whl/cu128 | |
| # # If use bitsandbytes with cuda | |
| # https://github.com/bitsandbytes-foundation/bitsandbytes/releases/download/continuous-release_main/bitsandbytes-1.33.7.preview-py3-none-manylinux_2_24_x86_64.whl | |
| # # If use llama-cpp-python with cuda | |
| # https://github.com/abetlen/llama-cpp-python/releases/download/v0.3.4-cu124/llama_cpp_python-0.3.4-cp311-cp311-linux_x86_64.whl | |
| # # If use llama-cpp-python CPU | |
| llama-cpp-python == 0.3.9 | |
| # If process file feature enable | |
| beautifulsoup4 == 4.13.4 | |
| requests == 2.32.3 | |
| langchain_chroma == 0.2.2 | |
| langchain_huggingface == 0.1.2 | |
| langchain_community == 0.3.19 | |
| chromadb == 0.6.3 | |
| pymupdf == 1.25.1 | |