How 0xzee/streamlit_langgraph_chatbot works
This is a Streamlit chat app backed by a LangGraph workflow and Groq inference. The user starts a session, selects a model and thread ID, submits a prompt, and sees the assistant response; the interface also displays chat history as JSON. The engine invokes the selected model and uses in-memory checkpointing for thread-scoped workflow state. Model selection can rebuild the engine. The README describes these capabilities; the graph reflects the supplied source, not unsampled implementation details.
Main entry points: chat_engine.py ChatBot engine · main.py Chat UI and sessions
Components 6
Chat interface
- Chat UI and sessions
main.py
Chat engine
- ChatBot engine
chat_engine.py - LangGraph workflow
chat_engine.py
Runtime integrations
- Memory checkpoint
chat_engine.py - Groq inference
Other components
- Chat user
Connections 9
- Chat user → Chat UI and sessions: uses
- Chat UI and sessions → ChatBot engine: creates
- Chat UI and sessions → ChatBot engine: updates model
- Chat UI and sessions → ChatBot engine: submits prompt
- ChatBot engine → LangGraph workflow: invokes
- LangGraph workflow → Groq inference: invokes model
- LangGraph workflow → Memory checkpoint: reads and writes
- LangGraph workflow → ChatBot engine: returns response
- ChatBot engine → Chat UI and sessions: returns result