Clef是来自Cloudflare的27B型多式联运决定模型，现在可在Ollama上找到。 用户可以在当地运行，也可以在Ollama的云上运行。 克莱夫将一个状态和一个打字问题的计划转化为决定。 它支持文本和图像输入，并与Jev和System One API兼容。
根据Apache 2.0许可，克莱夫从Qwen3.8-27B和发布进行微调。
要点
- 多式联运：就文本、JSON或图像作出决定，包括截图、收据、表格和照片。
- 一个传球：克莱夫读过一次输入，并同时给每个问题评分，而不是逐字写答字。
- 铅精度：克莱夫·超过·Cloudflare的决定指数，并在大部分基准测试套房上击败Jev.
你能建造的东西
Clef 支持下列使用案例：
- 提出请求的路线：确定目的地和每项请求何时适用。 Clef 返回选择的目的地和每个目的地的概率。
请检查access-date=中的日期值 (帮助) Clef返回真假和每个概率。
- 应用政策：定义规则，允许结果收到基于所提供文本或图像的打字决定。
- 阅读文件：提供一张页面和字段的照片，以评价有概率的实地决定。
- 工具调用节制：运行前检查agent的工具调用。
兹基准测试
Cloudflare宣布的结果
在Typesafe自己的工作流程evals上，克莱夫在4个领域中3个战胜Jev:
43 eval 基准测试的空闲度，以毫秒计：
更多信息见：
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Clef, a 27B multimodal decision model from Cloudflare, is now available on Ollama. Users can run it locally or on Ollama’s cloud. Clef turns a state and a schema of typed questions into decisions. It supports text and image input and is compatible with the Jev and System One APIs.
Clef is fine-tuned from Qwen3.8-27B and released under the Apache 2.0 license.
Highlights
- Multimodal: Make decisions over text, JSON, or images, including screenshots, receipts, forms, and photos.
- One pass: Clef reads the input once and scores every answer to every question at the same time, instead of writing a response word by word.
- Leading accuracy: Clef tops Cloudflare’s Decision Index and beats Jev on most of its benchmark suite.
What you can build
Clef supports the following use cases:
- Route a request: Define destinations and when each applies. Clef returns the chosen destination and each destination’s probability.
- Check a condition: Define a yes-or-no question and the evidence. Clef returns true or false and the probability of each.
- Apply a policy: Define rules and allowed outcomes to receive a typed decision based on supplied text or images.
- Read a document: Supply a photo of a page and the fields to evaluate for field-by-field decisions with probabilities.
- Tool-call moderation: Check an agent’s tool call before it runs.
Benchmarks
Results from Cloudflare’s announcement, scored on evaluations from the Jev Decision Index:
On Typesafe’s own workflow evals, Clef beats Jev in 3 of 4 areas:
Latency across 43 eval benchmarks, in milliseconds:
For more information, see:
Auto-fetched and translated by AI Atlas. All rights belong to the original author.