Chainlit — Build Conversational AI Apps Fast

Key Features

Instant Chat UI

Beautiful, responsive chat interface with zero frontend code. Streaming, markdown, code highlighting, and LaTeX.

File Upload & Display

Users can upload files. Agents can send images, PDFs, and data visualizations.

Feedback & Human Evaluation

Built-in thumbs up/down feedback on AI responses. Export for fine-tuning.

Multi-Step Display

Show intermediate "thinking" steps, tool calls, and chain-of-thought reasoning.

Authentication

Built-in auth with OAuth (Google, GitHub) or custom login.

Data Persistence

Conversation history saved automatically. Resume sessions across visits.

Framework Integration

# LangChain integration
@cl.on_message
async def main(message):
    result = await chain.ainvoke({"input": message.content},
        callbacks=[cl.LangchainCallbackHandler()])
    await cl.Message(content=result).send()

FAQ

Q: What is Chainlit? A: A Python framework for building conversational AI interfaces in minutes — chat UI, streaming, file upload, feedback, and auth included. 12K+ stars.

Q: How is Chainlit different from Streamlit? A: Chainlit is purpose-built for chat/conversational AI with streaming, multi-step display, and LLM framework integrations. Streamlit is a general data app framework.