Cloudflare previews Clairic, a System One extraction model
Original title: Introducing Clairic, a new System One class of extraction models
Clairic is hosted on Workers AI as @cf/cloudflare/clairic, charges only for input, and open weights are planned.
Cloudflare announced a preview of Clairic, which it calls the first System One extraction model. It extracts, computes, estimates, and forecasts typed values from text and up to four images in one forward pass, returning spans, numbers, dates, lists, or distributions with confidence scores, or null. Cloudflare says it does not generate tokens, so it avoids LLM-style mistyped answers, and that it intends to release the weights later.
- The model is available as @cf/cloudflare/clairic on Workers AI.
- Cloudflare says responses typically take under a second and incur no output-token charge.
- Outputs include extracted spans with character offsets and an 80% interval for estimates.
- Decision models Jev and Clef remain separate from this extraction class.
- Open weights are intended but not yet released.
- Vendors
- Cloudflare
- Products
- Workers AI
- Organizations
- TypeSafe
- AI models
- Clairic
From the original article
15 min read at blog.cloudflare.com
Add your teaser here. It should complement the headline, not repeat it. Teaser cannot be longer than 240 characters.
By Alex Reneau · blog.cloudflare.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.