Clef-flash
The fast 9B Clef for latency-critical, hot-path decisions.
Clef-flash is the 9B member of Cloudflare’s open-source Clef decision family, fine-tuned from Qwen3.5-9B for latency-critical decisions. Cloudflare reports a 38.8 ms median latency and top scores on BFCL among Clef models (vendor-reported). Apache-2.0 weights on Hugging Face; served on Workers AI.
- Provider
- Cloudflare
- Type
- Decision model
- Released
- Oct 1, 2026
- Input
- text, image
- Output
- decision
- Open weights
- Yes
- License
- Apache-2.0
Best for
- Extraction & classification
- Real-time / low latency
- High volume / low cost
Strengths
- ≈39 ms median latency (vendor-reported)
- Open weights (Apache-2.0)
More from Cloudflare
- Clef — Cloudflare. Cloudflare’s open-source 27B multimodal decision model for high-precision decisions.
- Clef Omni — Cloudflare. Clef’s mixture-of-experts member, deciding over text, images, audio and video.
Tools that use it
- Workers AI — Cloudflare. Serverless GPU inference on Cloudflare’s global network.