Add Support for Claude & Gemini APIs Alongside OpenAI

[michiosw]

Currently, the implementation assumes the use of an OpenAI API key. However, for users who prefer alternative models like Claude or Gemini, it's unclear how this would be handled. Supporting multiple LLM providers would enhance flexibility and allow users to choose the best model for their needs.

[zhangshaorui]

server.py

llm = ChatOpenAI(
# base_url='https://api.deepseek.com/v1',
# model = 'deepseek-chat',
# api_key = os.getenv('DEEPSEEK_API_KEY')
base_url='https://dashscope.aliyuncs.com/compatible-mode/v1',
model='qwen-max',
api_key=os.getenv('OPENAI_API_KEY')
)

[allanpk716]

server.py

llm = ChatOpenAI( # base_url='https://api.deepseek.com/v1', # model = 'deepseek-chat', # api_key = os.getenv('DEEPSEEK_API_KEY') base_url='https://dashscope.aliyuncs.com/compatible-mode/v1', model='qwen-max', api_key=os.getenv('OPENAI_API_KEY') )

thx，rebuild the docker image, it worked

[michiosw]

Hey @zhangshaorui , thank you for the support! We would greatly appreciate it if you could open a PR. From what I’m seeing, it’s using a proxy AI API provider that follows the OpenAI style — does this mean you could connect any provider?
Best,
Michel