Cloudflare Releases Clairic, a System One Extraction Model on Workers AI
By Asif Razzaq โ October 20, 2026
Cloudflare has released a preview of Clairic, a new System One extraction model on Workers AI. The Cloudflare Clairic extraction model reads text or images and returns typed values rather than generated text. It targets a job developers usually hand to LLMs: pulling totals, dates, and IDs into JSON. Clairic answers every requested field in a single forward pass, with no reasoning and no output tokens.
TL;DR
- Size: Parameter count not disclosed. The model offers a 65,536-token context window and accepts up to 4 images per request.
- Runs on: Cloudflare-hosted GPUs via Workers AI (
@cf/cloudflare/clairic). There is no local option until weights ship. - Performance: Sub-second extraction in Cloudflare's tests. No standard accuracy benchmark has been published yet.
- Best: In Cloudflare's claims pipeline, the extraction step took 0.46 s versus 16.1 s for GLM-5.3.
- Worst: 4 of 21 TSA next-day forecasts fell outside its 80% interval. Multi-step arithmetic remains unreliable.
What Is Clairic?
Clairic is a Cloudflare model that extracts, co[nverts, and returns structured...]
Editor's note: The source text appears to be truncated mid-sentence at the end of the "What Is Clairic?" section and after the TL;DR. The content above preserves the original information without fabricating the missing continuation; please supply the remaining text for a complete cleanup and restructure.
via MarkTechPost