Cloudflare Clairic and Clairic-flash
Open-weight document-extraction models that pull typed fields, tables, dates and line items from text, images and PDFs with confidence scores and character spans; Clairic is 27B, Clairic-flash is a lighter 9B.
Why it matters
Purpose-built extraction beats general LLMs on the vendor's internal evals (image-reading errors 207 to 0 vs preview) at $0.10/M input tokens with sub-second latency — and the weights are open under Apache 2.0, not just an API.
What you could build with it
A small accounting or legal-tech startup could build an invoice and contract intake service on Clairic-flash that extracts line items, totals and dates from messy PDF scans into structured records, charging per document processed instead of per LLM token.
Does it hold up?
Too early to judge: the eval gains are the vendor's own numbers and the models are a day old, but Apache-2.0 weights plus a hosted Workers AI path make trying it cheap.
Built with Cloudflare Clairic and Clairic-flash
- Cloudflare releases open-weight Clairic models for data extractionunite.ai · Independent coverage of the two models, pricing, and the open-weights release.
- Cloudflare/clairic on Hugging Facehuggingface · Model card for the 27B extraction model: Qwen3.8-27B base, typed fields with confidence scores, PyTorch/Transformers usage, Apache 2.0.
Learn more
First spotted on article: source.
More AI models
Edge0-145B: open 145B MoE that runs on a single consumer GPUEdge0 open-sourced 145B and 120B latent-MoE models derived from Qwen3.8-Flash-Next, plus an inference framework that…models · JEV 0.74Strata: run a 125B model on a gaming PCFree, MIT-licensed local inference engine that runs Qwen3.8-Flash-Next (125B MoE, about 6B active per token) on…models · JEV 0.73Qwen4-Max: Alibaba's new flagship model goes liveAlibaba's 2.4T-parameter flagship is now generally available via API, with a 1M-token context, native vision, and…models · JEV 0.67smolmaxxing: run 16B LLMs on a 1 GB TV box or a 2013 iPadAn MIT-licensed llama.cpp patch set that runs language models up to a 16B MoE on devices with about 1 GB of usable RAM…models · JEV 0.65Blue TTS v3: a 388MB open text-to-speech model with voice cloningAn open-source 388MB text-to-speech model (MIT) with voice cloning from about 10 seconds of speech, 7 languages, 15…models · JEV 0.65Haru-4BOpen full-duplex spoken-dialogue foundation model that listens and speaks simultaneously (7.5 Hz frames, 133 ms per…models · JEV 0.63
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.