{
  "id": 11234850,
  "title": "Clef: our open-source decision models",
  "url": "https://urgent.news/2026/10/01/clef-our-open-source-decision-models",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-01T16:18:57.000Z",
  "source": {
    "name": "Hacker News",
    "slug": "hacker-news",
    "url": "https://blog.cloudflare.com/clef-decision-models/"
  },
  "original_language": "en",
  "account": "Cloudflare has introduced two open-source decision models, Clef and Clef-flash, designed to provide rapid, accurate, and cost-effective AI-driven classifications for various applications. Unlike traditional classifier models, these models generate structured outputs, enabling seamless integration into workflows when decision-making is required. These models are capable of handling a wide range of input types, including text and images, without the need for constant retraining to incorporate new categories. Cloudflare's Clef model currently leads in performance, outperforming the Jev Decision Index in terms of accuracy and speed. The Clef model is available on Hugging Face under an Apache 2.0 license, allowing for local experimentation and fine-tuning. Cloudflare's unique decision model leverages image classification capabilities, a feature not present in competing models like Jev, which specializes in text classification. Additionally, Clef boasts a larger 64k context window compared to Jev's 32k, enabling more extensive input processing. By being hosted on Cloudflare's Workers AI infrastructure, Clef models benefit from edge-based GPU acceleration, resulting in reduced latency and faster decision-making. This edge deployment offers several advantages, including minimal network latency, quicker response times, and the ability to perform decisions within the path of traffic flow. Clef and Clef-flash are enterprise-ready, ensuring that customer data remains secure and private, with Cloudflare not reading, storing, or training on user requests or responses. The release of these models marks a significant step forward in the democratization of AI-driven decision-making, offering businesses and developers powerful tools to streamline their workflows and enhance their applications.",
  "summary": null,
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}