Cloudflare's Clairic reads invoices for $0.10 per million tokens

Cloudflare released a preview of Clairic on October 19, a model that reads documents such as invoices and returns typed values instead of written answers. It costs $0.10 per million input tokens, with no charge for output. If you pay a chatbot-style model to pull totals out of documents, this is a cheaper tool built for that job. Help Net Security also reported the release.

It fills fields instead of writing text

A request holds the text to read, plus up to four images, and the fields you want back, each with a type: text, number, date, duration, list, estimate or forecast. Clairic answers every field in one pass and generates no tokens, which is why Cloudflare charges only for input and says replies typically take under a second.

Each answer carries a confidence score and p_null, the model’s estimate that the value is missing. Missing values come back as null rather than a plausible guess. In Cloudflare’s own example, an email with no tracking number returned null for that field, even though the model’s next best guess was the tax amount, “$52.10”, at 0.42 confidence.

That matters because invoice extraction with a general language model fails quietly. The output looks right whether or not it is. A confidence number per field gives you a place to draw a line between “post it” and “check it”.

What it costs, worked out from Cloudflare’s numbers

Cloudflare’s model docs run a short sample invoice through six fields. The response reports 655 input tokens: 61 for the invoice text and 472 for the field definitions. So the field list, not the document, drove most of that bill.

At $0.10 per million tokens, a request like that costs $0.0000655, or about 66 cents for 10,000 of them. The Workers AI pricing page counts Clairic at 9,091 neurons per million input tokens and gives every account 10,000 neurons a day free, on both the Free and Paid plans. By our arithmetic that is about 1.1 million tokens a day, or about 1,680 requests the size of Cloudflare’s sample. Real invoices are longer than 61 tokens, so expect fewer.

In a test Cloudflare ran on a damaged-product claim, Clairic extracted the facts in 0.46 seconds against 16.1 seconds for the GLM-5.3 language model. These are the company’s own figures; we found no independent test.

The limits are in Cloudflare’s own docs

• No PDFs. Images must be PNG, JPEG or WebP, at most four, 4 MiB each, and remote URLs are not accepted. A scanned PDF has to become text or images first.
• Weak at long sums. Cloudflare says one- and two-step calculations look reliable, but longer chains and totals across many table rows do not. Its advice: “let code do the math.”
• Well formed is not correct. Cloudflare’s extraction guide says typed results “do not guarantee correctness” and asks for human review before “consequential or irreversible actions.”
• A preview with no guarantees. The model terms say Cloudflare “may cease providing it” without refund. Cloudflare says it intends to release open weights, which would ease that risk, but has not done so yet.

What to do now

1. If invoices or receipts arrive as text or photos, call @cf/cloudflare/clairic from a Worker or the REST API on the free allocation. An account and an API token are enough.
2. Test on 20 invoices you have already keyed in by hand, and count wrong fields per invoice.
3. Ask only for the values on the page, then compute totals and due dates in your own code.
4. Set a confidence cutoff and send everything below it, and every null, to a person before it reaches payments.

If you have not yet decided whether invoices are the job worth automating first, our one-week tally method answers that before you write any code.

Sources

• Introducing Clairic, a new System One class of extraction models, Cloudflare blog, October 19, 2026
• Clairic model reference, Cloudflare Docs
• Extraction models, Cloudflare Docs
• Clairic changelog entry, Cloudflare, October 19, 2026
• Workers AI pricing, Cloudflare Docs
• Cloudflare bets you’re overpaying an LLM to read your invoices, Help Net Security, October 20, 2026