Clef: Cloudflare's open-source decision models and RL fine-tuning platform
Cloudflare released Clef and Clef-flash, its first trained decision models — typed probability outputs, Jev-API compatible, Apache 2.0 weights, hosted on Workers AI — plus a reinforcement-learning fine-tuning service.
Why it matters
First major-cloud open decision model family: a drop-in alternative to hosted decision APIs with open weights, a vision encoder and 64k context, plus Cloudflare's own RL fine-tuning loop.
What you could build with it
An indie developer could put Clef in front of a support inbox to auto-route tickets with calibrated confidence scores, escalating only low-confidence cases to humans — a cheaper, deterministic gate than routing everything through a frontier chat model.
Does it hold up?
Cloudflare reports 43 internal benchmarks (top score on 7 of 10 decision benchmarks, 2.5-13x lower latency than Jev) and a real deployment on its Threat Intelligence team classifying domains in 2.2s vs 4.7s for a general LLM — vendor numbers so far; independent verification still pending.
Built with Clef
- Introducing Clef — Cloudflare changelog (Oct 1, 2026)cloudflare docs · Official changelog entry dated Oct 1, 2026: @cf/cloudflare/clef and @cf/cloudflare/clef-flash are available on Workers AI today, with Apache 2.0 weights open-sourced on Hugging Face.
- Clef: open-source decision model — Cloudflarecloudflare · Cloudflare's Clef product page describing the open-source decision models and sign-up for the RL fine-tuning design-partner program.
Learn more
First spotted on article: source.
More AI models
ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71DeepSeek V4-Flash official API: public beta with upgraded agent capabilities, Responses API and Codex supportDeepSeek's official V4-Flash is now a public beta API with massively upgraded agent capabilities - benchmark scores…models · JEV 0.69Mistral Forge: enterprise platform for training and continuously improving proprietary modelsMistral AI launched Forge (Sep 28) — a full-lifecycle model training platform (pre-training, SFT, DPO/ODPO, RL…models · JEV 0.69kev: open, trainable Jev-like family of small decision models on Qwen3.5/3.8A family of small decision models (0.8B to 27B) built on Qwen3.5/3.8 that reproduces Jev's prefill-only typed-decision…models · JEV 0.68GPT-Synopsys: OpenAI and Synopsys multi-year deal to build an AI model that operates EDA toolsOpenAI and Synopsys signed a multi-year agreement on Sep 30 to jointly develop GPT-Synopsys, a specialized model that…models · JEV 0.66
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.