@interjc: CF has released clef-omni, which can now recognize audio and video. If the decision model can achieve high cost-perform…
Summary
Cloudflare 发布 clef-omni，支持音频和视频输入，被用于 PII 检测、恶意域名识别等场景；同时 clef-flash 价格降至每百万输入 token 0.038 美元，具备较高性价比。
View Cached Full Text
Cached at: 10/10/26, 04:39 AM
CF has released clef-omni, which can now recognize audio and video. If the decision model can achieve high cost-performance, the demand is actually pretty substantial. I’m also evaluating handing over the audio recognition work using Gemini in some products to clef to give it a try.
Dane Knecht 🦭 (@dok2001): Clef-omni is out. We add audio and video input, plus text and images.
@Cloudflare teams are using it to find PII, detect malicious domains, and the docs repo to close spam issues.
Clef-flash is now cheaper at $0.038 per M input tokens.
Similar Articles
Introducing Clef-omni with full multimodality, plus a faster Clef and a cheaper Clef-flash (7 minute read)
Cloudflare发布开源权重决策模型Clef-omni，支持音频、视频、图像和文本全模态输入，并同时加快了Clef速度、降低了Clef-flash价格。模型基于Qwen3-Omni-30B-A3B-Instruct MoE架构构建，去掉了文本转语音输出组件。
Cloudflare released Clef-omni, a 30B-A3B decision model that takes audio and video directly (Apache-2.0, Qwen3-Omni base)
Cloudflare 发布 Clef-omni，一个基于 Qwen3-Omni-30B-A3B 后训练的多模态决策模型，可直接接收音频与视频输入，在单次前向传播中输出每个选项的概率，意图路由类基准表现强劲。
@julien_c: and... new clef drop from cloudflare!
Cloudflare 发布了新的多模态模型 Clef-Omni，这是一个基于 Qwen3-Omni-30B-A3B-Instruct 后训练的 30B-A3B 混合专家多模态模型，可将状态和结构化问题模式一次性转换为概率决策，无需自由文本生成。
Clef: our open-source decision models
Cloudflare 发布了两个开源决策模型 Clef 和 Clef-flash，托管于 Workers AI，采用 Apache 2.0 许可并在 Hugging Face 上开放下载，主打低成本、快速且一致的结构化输出，目前在 Jev Decision Index 榜单领先；同时推出新的强化学习微调平台，支持客户针对自身用例对 Clef 进行微调。
Clef
Cloudflare introduces Clef and Clef-flash, open-source decision models hosted on Workers AI for high-speed classification and agentic workflows, alongside a new reinforcement learning platform that lets developers fine-tune decision models with their own data.