Chris Chen

可以的，用cloudflare进行中转，在客户端这里用自定义域名代替api.openai.com。用官方sdk估计也可以

应该是openai.api_base

@zhayujie 测试成功。后面一定要加上`/v1/` `openai.api_base = 'https://自定义域名/v1/'`

> openai初始化的时候会去读环境变量OPENAI_API_BASE，可以尝试一下直接把这个环境变量设进docker看行不行： https://github.com/openai/openai-python/blob/75c90a71e88e4194ce22c71edeb3d2dee7f6ac93/openai/__init__.py

Same issue here

Got same error ``` loader = PyPDFLoader(fpath) pages = loader.load_and_split() text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000,chunk_overlap=20) docs = text_splitter.split_documents(pages) embeddings = OpenAIEmbeddings() rds = Redis.from_documents(docs, embeddings, redis_url="redis://localhost:6379", index_name='link') ```

> > Got same error > > > > ``` > > loader = PyPDFLoader(fpath) > > pages = loader.load_and_split() > > text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000,chunk_overlap=20) > > docs = text_splitter.split_documents(pages)...

same issue 0.4.31

Any update for this? 🙏

It's because connecters such as pptx or docx only support to build index one by one file?