cloudflare
Cloudflare: Clef
cloudflare/clef
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Cloudflare Clef (cloudflare/clef) is described in the model documentation as a 27B multimodal decision model post-trained from Qwen/Qwen3.8-27B. It accepts a textual or JSON state, plus optional images or video, and emits one logit for every allowed answer in a typed question schema in a single pass; a per-question softmax yields probabilities. OpenRouter separately catalogs it as text+image-to-decisions and lists its context length. All evaluations in this source pack appear in the vendor's model card, so workload validation remains important.[1][2]
Strengths
- The joint schema head is described as directing evidence to each typed question and scoring all allowed options jointly, avoiding free-form generation and output parsing.[2]
- A record can mix text or JSON state with optional media; the model card lists images and video, while OpenRouter confirms text and image inputs.[1][2]
- The card reports Jev/SystemOne API compatibility and supplies local-inference plus SGLang serving examples.[2]
Limitations
- It is a decision model, not a general text generator: outputs are option scores or probabilities for the submitted schema only.[2]
- Evidence is incomplete: OpenRouter lists text+image and no benchmark data, while video support and all performance figures in this pack come from Cloudflare's model card; its Decision Index run is described as internal.[1][2]
- The documentation gives tested examples on specific GPUs but does not establish general hardware requirements outside those cases.[2]
Where to use it
- Potentially suited to customer-message triage, such as routing to a team and assessing urgency, when questions can be fixed in advance.[2]
- Potentially suited to invoice or security-incident workflows that need structured actions rather than prose responses.[2]
- Potentially suited to image-assisted checks, such as receipt-legibility review, where a typed yes/no or choice output is sufficient.[2]
Sources
- [1] OpenRouter catalog · Cloudflare: ClefSources checked: Nov 13, 2026
- [2] Model documentation · Cloudflare: ClefSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Clef is Cloudflare's open-source 27B multimodal decision model, a fine-tune of Qwen3.8-27B served on Workers AI. It turns a state (text or structured JSON) plus a schema of typed questions...
Specifications
- Model ID
- cloudflare/clef
- Creator
- cloudflare
- Added to OpenRouter
- 01.10.2026
- Context length
- 65,537
- Modality
- text+image->decisions
- Input modalities
- text, image
- Output modalities
- decisions
- Token pricing
- No per-token price in the source snapshot — see OpenRouter for the billing unit
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | — |
| Coding | No measurement | — |
| Agents | No measurement | — |
| Writing | No measurement | — |
| Reasoning | No measurement | — |
| Document analysis | No measurement | — |
| Vision | No measurement | input: image |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| DeepInfrafp8 | 65,537 | — | — | — | — | 100% |
| PrimeIntellect | 65,536 | — | — | — | — | 100% |
| Cloudflare | 65,536 | — | — | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.