Scenarios
Concrete, copy-paste-ready walkthroughs for the things developers actually build: retrieval, chat, and classification against one TokenHub key.
Batch classification with token-hub
Run large-volume classification jobs through token-hub with aiohttp, bounded concurrency, retries, and the verified moonshot-v1-8k route.
- Python
- asyncio
- aiohttp
- CSV
Chat app with streaming responses
Build a server-side chat route that streams OpenAI-compatible output through token-hub using the currently verified moonshot-v1-8k route.
- Next.js 14
- React
- Edge Runtime
- Server-Sent Events
RAG answer synthesis through token-hub
Use token-hub as the generation step in a retrieval pipeline while keeping embeddings and vector storage in your own stack.
- Python
- LangChain
- Chroma
- OpenAI-compatible chat