{"author":"pinglin","children":[{"author":"efriis","children":[],"created_at":"2024-09-03T05:26:32.000Z","created_at_i":1725341192,"id":41431561,"options":[],"parent_id":41427135,"points":null,"story_id":41427135,"text":"Erick from LangChain here - would recommend either deploying with LangGraph Cloud or anywhere that deploys a FastAPI easily.<p>LangGraph Cloud: <a href=\"https:&#x2F;&#x2F;langchain-ai.github.io&#x2F;langgraph&#x2F;cloud&#x2F;\" rel=\"nofollow\">https:&#x2F;&#x2F;langchain-ai.github.io&#x2F;langgraph&#x2F;cloud&#x2F;</a>\nExample (chat-langchain): <a href=\"https:&#x2F;&#x2F;github.com&#x2F;langchain-ai&#x2F;chat-langchain?tab=readme-ov-file\">https:&#x2F;&#x2F;github.com&#x2F;langchain-ai&#x2F;chat-langchain?tab=readme-ov...</a><p>FastAPI example with LangServe (old chat-langchain implementation): <a href=\"https:&#x2F;&#x2F;github.com&#x2F;langchain-ai&#x2F;chat-langchain&#x2F;blob&#x2F;langserve&#x2F;backend&#x2F;main.py\">https:&#x2F;&#x2F;github.com&#x2F;langchain-ai&#x2F;chat-langchain&#x2F;blob&#x2F;langserv...</a><p>Or just use regular fastapi endpoints instead of langserve.add_routes with a fastapi server: <a href=\"https:&#x2F;&#x2F;fastapi.tiangolo.com&#x2F;\" rel=\"nofollow\">https:&#x2F;&#x2F;fastapi.tiangolo.com&#x2F;</a>","title":null,"type":"comment","url":null}],"created_at":"2024-09-02T17:16:46.000Z","created_at_i":1725297406,"id":41427135,"options":[],"parent_id":null,"points":2,"story_id":41427135,"text":null,"title":"Ask HN: How do you deploy Python-based LangChain or LlamaIndex code on prod?","type":"story","url":null}