Needle 3
edgeA retrieval model with 121M parameters, built to find the needle in the haystack: hand it a pile of documents and a question, and it embeds chunks so the right page surfaces first. Think of a search dog for your own files: small, tireless, and cheap to feed, at a scale that runs on a phone, or on an ESP32 microcontroller thanks to a community port. 36k downloads in its first days says the retrieval crowd noticed. Open weights.
- 120M
- 8k
- Cactus Compute
- Sep 2026
Save your hardware and every model page answers the real question: will it run on your machine, and how fast?
Join free - save your rig →Or run it in the cloud
No per-token API provider pricing tracked for Needle 3 yet. For flagship list prices, see the calculator.
Inference cost over time
Data accumulates from the first daily sync - longer ranges populate over time. Prices come from OpenRouter snapshots, not a historical API.