#Clefomni adds #audiototext and #videototext and #imageinputs via #Qwen3 Omni-30B-A3B-Instruct. #Clefflash is #cheaper than #Jev at $0.038 per M tokens with a 24k context window; #Clef is #faster via #SGLang. https://blog.cloudfl... #tech #news #ainews

The Clef model family is expanding with Clef-omni, which processes audio, video, images, and text in a single pipeline, alongside faster Clef inference and lower-priced Clef-flash.

#Clef #MultimodalAI #AI
https://blog.cloudflare.com/clef-faster-cheaper-multimodal/

the real difference between Clef and Gutsy

#jev #clef #opensource #huggingface

Cloudflare Clef decision models skip text generation to score options in as little as 38.8 ms, giving AI agents fast routing on Workers AI.

#Cloudflare #Clef #ClefFlash #AIAgents #WorkersAI #DecisionModel #OpenSourceAI #Qwen

#Cloudflare released #Clef and #Clefflash #decisionmodels, claiming faster, more accurate performance than #Jev via #Qwen backbones. Clef handles images and video, supports 64k context, costs $0.24 per million tokens, and is Jev-compatible on Workers AI or Hugging Face.…

Cloudflare lanza Clef, su modelo de decisión open source

Clef lanza modelos de decisión open-source y una plataforma de fine-tuning con RL: descubrí cómo funciona el Clef modelo de decisión y animate a probarlo.

#clef #cloudflare #workersai #modelosdedecisión #jev

Cloudflare shipped #Clef, open-weight decision #models that pick an answer instead of writing one. #Clef-flash decides in 38.8 ms, reads photos, free to ~1.2M tokens/day. Catch: no GGUF has the decision head, no laptop runs it yet. #AI