Cloudflare releases Clef, open decision models that take Jev's requests

TL;DR

• On 1 October Cloudflare released two open decision models, Clef and Clef-flash, built on Qwen and licensed under Apache 2.0.
• They take the same requests as TypeSafe AI's Jev, return a probability for every allowed answer instead of text, and read images.
• In Cloudflare's own runs Clef-flash has a 38.8 ms median latency against Jev's 524.1 ms; The Register puts Clef at $0.24 per million tokens against Jev's $0.042.