{"schema":"postcutoff/event@1","as_of":"2026-10-20T07:28:00+02:00","url":"https://postcutoff.com/e/2026-10-19-cloudflare-clairic-extraction-model/","md":"https://postcutoff.com/e/2026-10-19-cloudflare-clairic-extraction-model/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-10-19-cloudflare-clairic-extraction-model","date":"2026-10-19","date_precision":"day","short_title":"Cloudflare launches Clairic, a 'System One' model that extracts, computes and forecasts values in one forward pass without generating tokens","deck":null,"takeaway":"In Cloudflare's insurance-claims demo a Clairic → Clef → GLM-5.3 pipeline ran in 19.3 s at $6,032 per million claims, against 61.7 s and $11,740 for GLM-5.3 at every step.","category":"model-release","category_label":"Model releases","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"Cloudflare blog: Clairic, a System One extraction model","url":"https://blog.cloudflare.com/clairic-extraction-model/","type":"official","group":"primary","domain":"blog.cloudflare.com"},{"n":2,"title":"Cloudflare on X: Introducing Clairic","url":"https://x.com/Cloudflare/status/2112244042658423149","type":"official","group":"primary","domain":"x.com"},{"n":3,"title":"Michelle Chen on X: 'a brand new model paradigm'","url":"https://x.com/michellechen/status/2112241006104244584","type":"discussion","group":"reactions","domain":"x.com"}],"official":2,"filed":"2026-10-20","updated":"2026-10-20","orgs":["Cloudflare"],"title":"Cloudflare launches Clairic, a 'System One' model that extracts, computes and forecasts values in one forward pass without generating tokens","summary":"On Oct 19, 2026 Cloudflare released Clairic in preview on Workers AI, a \"System One extraction model\" that reads unstructured input in a single forward pass and returns typed values, text spans with offsets, null probabilities and confidence scores, without generating output tokens or reasoning step by step. In Cloudflare's insurance-claims demo a Clairic → Clef → GLM-5.3 pipeline ran in 19.3 s at $6,032 per million claims, against 61.7 s and $11,740 for GLM-5.3 at every step. Open weights are promised \"soon\".","key_facts":["Access: preview on Workers AI as @cf/cloudflare/clairic (env.AI.run() in a Worker or the REST API); billed for input tokens only, rate not stated","Four tasks: extract values, compute values from the input, estimate values, and forecast future values","Outputs: typed values, character-offset spans, a probability that the field is null, and a confidence score","Claims demo (Cloudflare): Clairic → Clef → GLM-5.3 took 19.3 s end to end vs 61.7 s for GLM-5.3 at every step; $6,032 vs $11,740 per million claims; Clairic's repair estimate was identical across runs while GLM-5.3's varied; Gemma 4 and Qwen reportedly failed the all-LLM version","Forecasting demo: TSA checkpoint counts forecast with 3.3% average error; 17 of 21 actual counts fell inside the 80% interval","No parameter count, architecture details or standard benchmark scores published; open weights planned with no date","A launch post by @michellechen on X (~183k views) called it 'a brand new model paradigm'"],"key_numbers":[],"tags":["extraction","structured-output","workers-ai","non-generative","forecasting"],"science":null,"body_md":"## What happened\n\nCloudflare added a second non-chat model family to Workers AI, after its Clef decision models of Oct 1. Clairic is built for\npulling fields out of documents, emails or forms and returning numbers with calibrated confidence, rather than writing text.\nBecause it produces no output tokens, Cloudflare charges only for input tokens.\n\nThe comparisons come from Cloudflare's own demos, not from standard benchmarks.\n\n## Why it matters\n\nIt is a bet that many enterprise \"LLM\" jobs (extraction, scoring, simple forecasting) are better done by fast, deterministic\nmodels that report uncertainty, with a general LLM called only for the parts that need reasoning.","disputed":[],"related":[{"id":"2026-10-01-cloudflare-clef-decision-models","url":"https://postcutoff.com/e/2026-10-01-cloudflare-clef-decision-models/","date":"2026-10-01","date_precision":"day","short_title":"Cloudflare releases Clef and Clef-flash, open-weight multimodal decision models that are Jev-API compatible","deck":null,"takeaway":"It is the second big-company follower in two days, after AWS's Strands Decider 2B.","category":"open-source","category_label":"Open source","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":8,"official":6,"filed":"2026-10-02","updated":"2026-10-10","orgs":["Cloudflare"]}],"people":[],"posts":[{"id":"x-cloudflare-2112244042658423149","title":"“Introducing Clairic, from Cloudflare. A System One extraction modeI, capable of…”","url":"https://postcutoff.com/p/x-cloudflare-2112244042658423149/"},{"id":"x-michellechen-2112241006104244584","title":"“a brand new model paradigm! introducing 𝚌𝚕𝚊𝚒𝚛𝚒𝚌: a system one model for…”","url":"https://postcutoff.com/p/x-michellechen-2112241006104244584/"}],"videos":[],"models":[{"id":"clairic","name":"Clairic","url":"https://postcutoff.com/m/clairic/"}],"changes":[{"date":"2026-10-20","type":"filed","text":"Created (Techmeme; Cloudflare blog and X posts read)"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-20","run":null,"sources_read":"Techmeme; Cloudflare blog and X posts read","updated":"2026-10-20","human_review":null,"version":null},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":172,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":111,"in_training_data":false},{"model_id":"gemini-4-argon","name":"Gemini 4 Argon","cutoff":"2026-03","days_after":202,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":141,"in_training_data":false}],"short_url":null}