Cloudflare Releases Clef and Clef-flash: Open-Weight Decision Models That Return Typed Probabilities Instead of Text

Cloudflare released Clef and Clef-flash, its first models trained by the Workers AI team, as open-weight decision models under Apache 2.0 that return typed probabilities instead of free-form text. Clef is post-trained from Qwen3.8-27B and Clef-flash from Qwen3.5-9B, both supporting noul, choice and score question types with up to 64 questions and 4 images per Workers AI request, and both are compatible with TypeSafe AI's Jev API. On Cloudflare's 10-benchmark Decision Index 0.2.1 shortlist a Clef model scored highest on 7, including BANKING77 macro-F1 of 94.20 versus Jev's 79.74, though Jev leads GPQA Diamond 78.3 to 48.0; all numbers are vendor-reported with no independent replication yet.

Cloudflare has released Clef and Clef-flash https://blog.cloudflare.com/clef-decision-models/ , the first models trained by its Workers AI team. They are decision models, not chatbots. Each reads an input state and a schema of typed questions. It returns a probability for every allowed answer, with no free-form text. Both are open-weight under Apache 2.0 and compatible with TypeSafe AI’s Jev API. Is it deployable? Yes, Both models run today on Workers AI https://developers.cloudflare.com/workers-ai/models/clef/ , and the weights are on Hugging Face https://huggingface.co/Cloudflare/clef for self-hosting. What a Decision Model Does An LLM generates tokens one at a time, and its output still needs parsing. A decision model only answers a fixed set of questions about an input. Clef supports 3 question types: - noul : yes/no, returns the probability of yes. - choice : picks 1 named option, with per-option probabilities and a confidence value. - score : rates against an ordered rubric, returning a probability-weighted score. On Workers AI, 1 request carries up to 64 questions and up to 4 images. TypeSafe AI launched Jev https://typesafe.ai/blog/introducing-system-one-models-and-jev , its first ‘System One’ model, on September 15, 2026. Open alternatives like Kev-9B https://huggingface.co/jaredpalmer/kev-9b and Laya https://huggingface.co/convaiinnovations/laya followed. Clef uses the same System One API. Switching from Jev means changing the endpoint and model name. How Clef Works Clef is post-trained from Qwen3.8-27B https://huggingface.co/Qwen/Qwen3.8-27B , and Clef-flash https://huggingface.co/Cloudflare/clef-flash from Qwen3.5-9B https://huggingface.co/Qwen/Qwen3.5-9B . Both keep the backbone’s vision encoder. Inference has 2 stages. The backbone first runs a single prefill-only pass over the state and questions. A small transformer, the joint schema head, then reads the final hidden states. It routes evidence to each question, lets fields cross-attend, and scores all options jointly. A per-question softmax turns logits into probabilities. Training froze both backbones and jointly optimized the routing head with rank-256 low-rank adapters. The loss pairs label-smoothed cross-entropy with a Brier loss for calibration. A secondary objective, Reinforcement Learning for Calibrated Decisions RLCD , gives partial credit to adjacent ordinal choices. Interactive Explainer Benchmarks: Where Clef Wins and Where It Does Not On Cloudflare’s 10-benchmark shortlist from the Decision Index https://clef-evals.workers-ai-mle.workers.dev/ 0.2.1 suite, a Clef model scored highest on 7. - BANKING77 macro-F1 : Clef 94.20 vs Jev 79.74. - CLINC150+OOS macro-F1 : Clef 97.43 vs Jev 89.27. - Home appliances case exact : Clef-flash 97.73 vs Jev 52.27. Jev keeps clear leads elsewhere. The full model card https://huggingface.co/Cloudflare/clef shows Jev ahead on GPQA Diamond 78.3 vs 48.0 . It also leads MMLU-Pro 82.7 vs 65.9 and BBH 92.9 vs 73.7 . On TypeSafe’s own workflow evals https://evals.typesafe.ai/ , Clef beat Jev in 3 of 4 areas, by small margins. Invoice processing was 64.7 vs 61.8, customer service 76.3 vs 76.0, and security incidents 62.9 vs 61.7. Jev leads agent trace observability, 71.6 vs 68.5. In Cloudflare’s threat intelligence workflow, Clef classified a domain in 2.2 seconds. gpt-oss-120b took 4.7 seconds. All numbers are vendor-reported, with no independent replication yet. Feature Comparison | Feature | Clef | Clef-flash | Jev | Kev-9B | Laya | |---|---|---|---|---|---| | Developer | Cloudflare | Cloudflare | TypeSafe AI | Jared Palmer | Convai Innovations | | Size | 27B | 9B | Not disclosed | 9B + 45.4M LoRA | 421M | | Backbone | Qwen3.8-27B | Qwen3.5-9B | Not disclosed | Qwen3.5-9B-Base | ModernBERT-large | | Weights | Apache 2.0 | Apache 2.0 | Hosted API | Apache 2.0 | Apache 2.0 | | Image input | Yes | Yes | No | No | No | | Context | 65,536 | 65,536 | 32K per Cloudflare | 65,536 8,192 validated | 512 English | | Median latency | 209.3 ms | 38.8 ms | 524.1 ms | 51.4 ms | 5.8 ms | | Hosted price input | $0.24/M | $0.09/M | $0.042/M | Self-host | Self-host | Cloudflare’s internal Decision Index run. All 5 implement the System One API. Sources: Clef docs https://developers.cloudflare.com/workers-ai/models/clef/ , Clef-flash docs https://developers.cloudflare.com/workers-ai/models/clef-flash/ , Clef card https://huggingface.co/Cloudflare/clef , Jev post https://typesafe.ai/blog/introducing-system-one-models-and-jev , Kev-9B card https://huggingface.co/jaredpalmer/kev-9b , Laya card https://huggingface.co/convaiinnovations/laya . Deployment and Fine-Tuning Both models are callable through the Workers AI binding env.AI.run , the REST API, or AI Gateway https://developers.cloudflare.com/ai-gateway/ . For self-hosting, the model cards list testing on a single H200 with BF16 weights. Cloudflare also announced a reinforcement learning service for tuning Clef on private data. It starts with Cloudflare’s forward-deployed engineers, with a self-serve platform later. The pipeline combines AI Gateway, Workers AI, Containers and a new Trainer component. Teams can apply via the design partner form https://www.cloudflare.com/resource/clef-rl-interest . Key Takeaways - Clef 27B and Clef-flash 9B are Apache 2.0, Jev-compatible decision models. - Median latency: 209.3 ms for Clef, 38.8 ms for Clef-flash, 524.1 ms for Jev. - Clef reads text, JSON, images and video within a 64K-token context window. - Jev still leads on knowledge-heavy tests like GPQA Diamond, MMLU-Pro and BBH. - An RL fine-tuning service starts with Cloudflare’s forward-deployed engineers. Check out the Model weight https://huggingface.co/Cloudflare/clef , Demo https://clef-evals.workers-ai-mle.workers.dev/ and Technical details https://blog.cloudflare.com/clef-decision-models/ . All credit goes to the researcher of this project. Also, feel free to follow us on Twitter https://x.com/intent/follow?screen name=marktechpost and don’t forget to join our 150k+ML SubReddit https://www.reddit.com/r/machinelearningnews/ and Subscribe to our Newsletter https://magic.beehiiv.com/v1/f5e63dd4-5653-4f09-83e2-321a8b1ba526?email={{email}} . Wait are you on telegram? now you can join us on telegram as well. https://t.me/machinelearningresearchnews Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us https://forms.gle/MJjjVDPS7whH8Ngs6 Asif Razzaq is the CEO of Marktechpost AI Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences.