Skip to content
Trending storyDeveloping

OpenRouter and SGLang make Cloudflare's Clef decision models available

2 articles2 sourcessince Oct 7Last article 1h ago ·

Overview

AISummary of 2 articles

OpenRouter has made Cloudflare's open-source Clef (27B) and Clef Flash (9B) decision models available, according to an OpenRouter post on X. The models return typed answers with probabilities instead of generated text.

OpenRouter lists input pricing at $0.24 per million tokens for Clef and $0.09 per million for Clef Flash, and says output is free. These are OpenRouter's own claims, not independently verified.

SGLang's X account says Cloudflare now serves Clef through SGLang's existing /v1/systemone API. SGLang says it contributed Clef support and performance optimizations upstream and validated them against the official reference implementation. It says Clef-Flash reaches sub-40 ms decision latency on a single H200 and answers 64 questions in one prefill, and that Clef-Omni support is coming next.

Written by AI from the articles below · updated Oct 12, 1:08 PM ET

Check the sources:

Latest developmentOct 12, 12:50 PM ET
Cloudflare serves Clef through SGLang's existing /v1/systemone API

Developments

2 developments

  1. Oct 12, 12:50 PM ET · 1 article
    Cloudflare serves Clef through SGLang's /v1/systemone API with Clef support upstreamed
    SGLang: Cloudflare serves Clef through SGLang's existing /v1/systemone API
  2. Oct 7, 2:17 PM ET · 1 article
    Cloudflare Clef decision models available on OpenRouter
    OpenRouter: Cloudflare's Clef decision models now available on OpenRouter

Article timeline

The articles in this story. Times are ET.

Oct 12
  1. SGLangOfficial
    Cloudflare serves Clef through SGLang's existing /v1/systemone API

    AISGLang announces that Cloudflare now serves the Clef model through SGLang's existing /v1/systemone API. The team says it contributed Clef support and performance optimizations upstream, validated against the official reference implementation. Clef-Flash reaches sub-40 ms decision latency on a single H200, answers 64 questions in one prefill with text and image support, and Clef-Omni support is coming next.

Oct 7
  1. OpenRouterOfficial
    Cloudflare's Clef decision models now available on OpenRouter

    AICloudflare's open-source Clef (27B) and Clef Flash (9B) decision models are available on OpenRouter. They accept text, JSON, or images and return typed answers with probabilities rather than generated text. Pricing is $0.24 per M input tokens for Clef and $0.09 per M for Clef Flash, with output free.

Heat trend

Not enough continuous observations to show a trend yet.