Coverage & methodology
What the archive actually holds, published so nobody has to take our word for it. Coverage is partial by design, and the shape of the gaps is stated rather than implied.
- Providers
- 6
- Models tracked
- 421including unpriced
- Observations stored
- 19,897append-only
- Published this run
- 820
- Archive reaches back to
- 2023-09-20
The documents
These are the normative texts, reproduced verbatim. Where a rule has a known hole, it is stated in the same paragraph as the rule.
- METHODOLOGY — how a number is produced: sources and trust order, what
verifiedandconfirmedeach mean, how conflicts resolve, what we will not publish, and what happens when we get one wrong. source - DISCLAIMER — what we commit to, what we do not guarantee, and how to check a value before you rely on it. source
- TERMS — the longer agreement. source
What the archive holds
| Published run | #18 |
|---|---|
| Run finished | 2026-08-16T18:06:22+00:00 |
| Oldest observation | 2023-09-20T20:39:22+00:00 |
| Newest observation | 2026-08-16T18:02:04+00:00 (2 hours ago) |
| Providers | 6 |
| Models | 421 |
| Observations stored | 19,897 |
| Values published | 820 |
| Of those, unconfirmed | 178 |
| Published basis | USD per 1,000,000 tokens |
| Scope | text and chat completion tokens only |
Observations by reader
A reader is either a deterministic parser written for one specific page, the language-model pass, or the third-party git backfill. Counts are over every observation ever stored, not just the published run.
| Reader | Observations | What it is |
|---|---|---|
unknown | 7,231 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
openrouter_json | 5,544 | Deterministic parser over the OpenRouter models JSON feed. No language model reads that source. |
llm | 3,642 | Language-model pass at temperature 0, read twice. Forbidden by prompt from inferring, estimating or converting units, and it cannot confirm a value the page's own parser did not emit. |
litellm_git_backfill | 2,930 | A public git history of a third-party pricing file. Never verified by us — it extends the series backwards to 2023 and is flagged wherever it appears. |
openai_astro_props | 184 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
gemini_pricing_tables+llm | 167 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
anthropic_pricing_table+llm | 60 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
gemini_pricing_tables | 56 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
zhipu_pricing_table+llm | 32 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
xai_models_blob | 27 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
openai_astro_props+llm | 15 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
deepseek_pricing_table+llm | 8 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
xai_models_blob+llm | 1 | Deterministic parser written for one specific page. It must prove its own coverage before its output is used at all. |
Sources behind the published run
Where the currently published values came from, and how many of them each source is responsible for.
| Source | Published values | Unconfirmed | URL |
|---|---|---|---|
| OpenAI API pricing (docs) | 274 | 75 | source pageplatform.openai.com |
| Gemini API pricing (docs) | 259 | 76 | source pageai.google.dev |
| OpenRouter models API | 184 | — | source pageopenrouter.ai |
| Z.ai GLM pricing (docs) | 53 | 21 | source pagedocs.z.ai |
| Claude API pricing (docs) | 26 | 6 | source pagedocs.claude.com |
| xAI models and pricing (docs) | 16 | — | source pagedocs.x.ai |
| DeepSeek API pricing (docs) | 8 | — | source pageapi-docs.deepseek.com |
Standing limits
Known and unresolved. The full table is in METHODOLOGY §9; these are the ones that change how you should read this site.
- Some published values are explicitly unconfirmed. 178 of 820 published values could not be corroborated under METHODOLOGY §3. They are shown with their full provenance precisely so you can see that they are unconfirmed, and they are barred from raising a change event.
- Some values are seen by only one source. When nothing corroborates a value, it is exactly as strong as the single source behind it.
- We do not warrant a maximum age for any value.
verified_attells you how old a reading is; it does not promise the reading will be refreshed within any interval. - Coverage is text and chat completion token pricing only. Embeddings, image generation, audio, moderation, fine-tuning surcharges and tool or web-search add-on fees are out of scope and excluded at extraction time. Not every vendor is covered, and not every model a vendor sells is covered.
- There is no notification mechanism. A correction is labelled, never announced. If you cite a number in a context where a later correction would matter, retain what you cited and re-check it.