Langchain-Chatchat/requirements_lite.txt
liunux4odoo b51ba11f45
支持通过配置项同时启动多个模型,将Wiki纳入samples知识库 (#2002)
新功能:
- 将 LLM_MODEL 配置项改为 LLM_MODELS 列表,同时启动多个模型
- 将 wiki 纳入 samples 知识库

依赖变化:
- 指定 streamlit~=1.27.0。1.26.0会报rerun错误,1.28.0会有无限刷新错误

修复优化:
- 优化 get_default_llm_model 逻辑
- 适配 Qwen 在线 API 做 Embeddings 时最大 25 行的限制
- 列出知识库磁盘文件时跳过 . 开头的文件
2023-11-09 22:15:52 +08:00

63 lines
1.1 KiB
Plaintext

langchain>=0.0.329 # 推荐使用最新的Langchain
fschat>=0.2.32
openai
# sentence_transformers
# transformers>=4.35.0
# torch>=2.0.1
# torchvision
# torchaudio
fastapi>=0.104.1
python-multipart
nltk~=3.8.1
uvicorn~=0.23.1
starlette~=0.27.0
pydantic~=1.10.11
unstructured[docx,csv]>=0.10.4 # add pdf if need
python-magic-bin; sys_platform == 'win32'
SQLAlchemy==2.0.19
faiss-cpu
# accelerate
# spacy
# PyMuPDF==1.22.5 # install if need pdf
# rapidocr_onnxruntime>=1.3.2 # install if need pdf
requests
pathlib
pytest
# scikit-learn
# numexpr
# vllm==0.1.7; sys_platform == "linux"
# online api libs
zhipuai
dashscope>=1.10.0 # qwen
# qianfan
# volcengine>=1.0.106 # fangzhou
# uncomment libs if you want to use corresponding vector store
# pymilvus==2.1.3 # requires milvus==2.1.3
# psycopg2
# pgvector
numpy~=1.24.4
pandas~=2.0.3
streamlit~=1.27.0
streamlit-option-menu>=0.3.6
streamlit-antd-components>=0.1.11
streamlit-chatbox==1.1.11
streamlit-aggrid>=0.3.4.post3
httpx~=0.24.1
watchdog
tqdm
websockets
# tiktoken
einops
# scipy
# transformers_stream_generator==0.0.4
# search engine libs
duckduckgo-search
metaphor-python
strsimpy
markdownify