Aryan Chauhan

Aryan Chauhan

Student
#84058360 followers1 following
☀️ 2 day streak
All activity
Aryan ChauhanhuntedWinnow
Winnow compresses RAG prompts before they hit your LLM, cutting token costs 50%+ while preserving meaning. Uses question-guided filtering + LLMLingua-2 for semantic accuracy. Key features: • FastAPI server with OpenAI-compatible proxy • Batch compression API • Question-aware filtering keeps answer-relevant tokens • Docker self-hosting, pip-installable SDK • MIT licensed
Winnow
WinnowKeep the signal. Drop the noise.
Aryan Chauhanleft a comment
🚀 Winnow is live (beta)! RAG pipelines eating your token budget? Winnow compresses prompts 50%+ while keeping answer-relevant tokens using question-guided filtering + LLMLingua-2. Try the live demo → https://trywinnow.vercel.app GitHub → https://github.com/itsaryanchauh... Currently testing/fixing along the way - feedback welcome! What would make this better for your stack?
Winnow
WinnowKeep the signal. Drop the noise.