{"rewrite":{"id":"r_5452f62287d29741816983a9","clusterId":"c_b39520ae4f33a1b879730af3","slug":"cloudflare-s-clairic-reads-invoices-without-generating-text","model":"deepseek-v4-1-flash","headline":"Cloudflare's Clairic Reads Invoices Without Generating Text","summary":"Cloudflare announced Clairic on October 19, 2026, an extraction model it calls a System One class tool. Instead of generating text word by word like an LLM, Clairic analyzes input in a single pass and outputs the requested values directly, including confidence levels and an 80% prediction range for estimates. It handles text and photographed images, returns null when information is absent, and computes derived figures such as outage duration.","whyItMatters":"Extraction at low cost and a null for missing fields is the part an LLM pipeline struggles with, and Cloudflare is pitching that gap rather than another general model.","webCardHtml":"\u003cp\u003eClairic answers in JSON, carries a confidence level on each extraction, and returns the value null when the requested field is not in the input, which Cloudflare says cuts the problem of filling in information that does not exist. In a worked example it read a customer failure inquiry, then calculated how long the failure had run and estimated affected users and lost orders in about 0.55 seconds.\u003c/p\u003e\u003cp\u003eCloudflare classifies it as an extraction model, separate from the decision models Jev and Clef, which pick from prepared options. The company calls Clairic the first model to expand System One to information extraction and numerical prediction.\u003c/p\u003e","blueskyPost":"Clairic returns null when the answer is not in the image, a refusal rule most extraction tools leave implicit. Cloudflare is betting that a reliable gap matters as much as the value.","twitterPost":"Clairic outputs null when the answer is not in the document. Reliable absence is the product.","threadsPost":"Clairic's confidence levels and an 80% prediction range are the part worth watching. Cloudflare is shipping a model that says how sure it is, so downstream systems can decide what to accept instead of trusting every field.","newsletterBlurb":"Cloudflare announced Clairic, an extraction model it groups under the System One label. It analyzes input in one pass and outputs requested values directly, returning JSON with confidence levels and null when information is absent. Cloudflare calls it the first model to expand System One to information extraction and numerical prediction.","attributionJson":"[{\"source\":\"GIGAZINE\",\"url\":\"https://gigazine.net/news/20261020-cloudflare-clairic/\",\"title\":\"Cloudflare announces \\\"Clairic,\\\" an AI that extracts needed information from text and images quickly and at low cost, and also handles calculations and predictions\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4617,"outputTokens":565,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1792483633,"createdAt":"2026-10-20T08:02:13.000Z","publishedAt":"2026-10-20T08:06:07.000Z","updatedAt":"2026-10-20T08:06:07.000Z"},"cluster":{"id":"c_b39520ae4f33a1b879730af3","canonicalTitle":"Cloudflareが文章や画像から必要な情報を高速・低コストで抽出するAI「Clairic」を発表、計算や予測にも対応","representativeArticleId":"a_72cf50763e2b9bdfc43375fe","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[\"Clairic\"],\"studios\":[\"Cloudflare\"],\"people\":[],\"type\":\"announcement\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-10-20T06:50:00.000Z","lastSeenAt":"2026-10-20T06:50:00.000Z","updatedAt":"2026-10-20T08:06:06.000Z"},"attribution":[{"source":"GIGAZINE","url":"https://gigazine.net/news/20261020-cloudflare-clairic/","title":"Cloudflareが文章や画像から必要な情報を高速・低コストで抽出するAI「Clairic」を発表、計算や予測にも対応"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":["Clairic"],"studios":["Cloudflare"],"people":[],"type":"announcement","domain":"other","is_roundup":false},"keyFacts":null}