AI Rate Card

AI Rate Card

AI Rate Card (airatecard.com) is an independent catalog of AI model offers and prices, read daily from the providers' own pricing pages and files. Every…

AI & machine learning#machine-learning

Fork copies the collection into your workspace, without the publisher's secrets, so you can add your own key and send requests. Star saves it to your starred APIs (Account › Marketplace) and helps others find it.

Overview

AI Rate Card (airatecard.com) is an independent catalog of AI model offers and prices, read daily from the providers' own pricing pages and files. Every answer is {"data": ..., "meta": {...}}; meta.cite says how to quote it (name AI Rate Card, link the card URL, give the checked date). Prices are USD per 1M tokens unless price_unit says otherwise. Without a key a little is allowed, more with a free personal key (the account, "API keys": send it as X-API-Key or Authorization: Bearer am_...); OAuth 2.1 tokens from the site's own server work the same and, with the scope provider, act for a provider (/v1/provider/*). Over a limit the answer is 429 with Retry-After. Guide for agents: /en/docs#agents, /auth.md, /llms.txt; the MCP server /mcp has the same data as tools.

Endpoints 32

catalog

  • GETSeller Analytics/v1/seller/:seller/analyticsThe provider's own analytics (free since 2026-10-16; was premium, Art 2026-10-08): demand_daily rows of its page and offers.
  • POSTSeller Requests Post/v1/seller-requestsAsk for a seller to be added to the catalog: its name, site and price page. Anyone may ask; the request is reviewed by people, nothing is published…
  • GETOffers/v1/offersThe catalog: one row per offer (one model at one provider), filtered and sorted by any of the parameters below, priced offers first. Prices in USD per 1M…
  • GETModels/v1/modelsThe catalog folded by model: one row per model with its number of providers, the lowest and median prices and the longest context.
  • GETFacets/v1/facetsdata = Catalog.facets (prices in USD); price limits are read in cur.
  • GETModel History/v1/models/:model_id/historyOne model's market by day (min, median, max price in USD, sellers), one model a request: days back from today, 90 at most (a larger number counts as 90).
  • GETModel/v1/models/:model_idOne model's passport and every provider's current price of it, cheapest first.
  • GETClasses List/v1/classesModel classes S/A/B/C/D of one direction (lead 2026-11-07, /classes): every base model with its letter (null = one source, score only), score 0-100, sources…
  • GETV1 Articles/v1/articlesThe site's articles (price comparisons written from the catalog), newest first, in the language asked for.

analytics

  • GETAnalytics Cheapest/v1/analytics/cheapest?task=textThe n models with the lowest of their own lowest current price of task.
  • GETAnalytics Dearest/v1/analytics/dearest?task=textThe same, highest first.
  • GETAnalytics Median/v1/analytics/median?task=textThe median of every qualifying model's own lowest price of task, and the n models closest to it.

workload

  • GETWorkload Rank/v1/workloadThe cheapest offers for a whole monthly workload (Art 2026-10-11, workload.py): a preset (chatbot, coding_agent, rag, summarization, long_context, vision,…
  • GETWorkload Presets/v1/workload/presetsThe ready-made workloads of /v1/workload (and the MCP tool cheapest_for_workload).
  • GETCompare Models/v1/compare2..4 models side by side (Art 2026-10-05), computed on request from the current offers: per field the value of each model, the best one and how far the others…
  • GETCost/v1/costThe cost of a workload at every provider of one model, computed on request from the current offers, cheapest first: model (id), input_tokens, output_tokens,…

provider

  • GETCheck Domain Route/v1/provider/check-domainPublic, no key: the catalog key a provider on this domain gets, whether the catalog already lists a seller on it (then claim_url - prove it is yours instead…
  • POSTValidate Route/v1/provider/validatePublic, no key: check a models.yaml text before publishing it on your domain. Every problem names the line, the model and the field, with a hint. Nothing is…
  • GETApplications List/v1/provider/applicationsThe applications (new providers) of the caller's organizations, newest first.
  • POSTApplications Post/v1/provider/applicationsStep 1: start listing a new provider - its name and domain. Returns the application with application_id, the domain-proof token and next_action. 200 when this…
  • GETApplication Get/v1/provider/applications/:ob_idThe application: every step with its status, what to publish where, the file problems of the last check and next_action.
  • POSTVerify Domain/v1/provider/applications/:ob_id/verify-domainStep 2: check the domain-proof token now (a file under /.well-known, a DNS TXT record or a meta tag). Optionally switch the way first.
  • POSTCheck Files/v1/provider/applications/:ob_id/check-filesStep 3: read models.yaml from the domain (or from models_url) and check it like the daily collection does; every problem with line, model and field.
  • POSTPublish/v1/provider/applications/:ob_id/publishStep 4: the file is read once more and the provider is queued for the collector (every few minutes); poll the application until status is published.
  • GETSeller Get/v1/provider/sellers/:keyA seller the caller's organization has verified: its card, the file read last (time, ok, error), its policy and next_action.
  • GETSeller Offers Get/v1/provider/sellers/:key/offersThe seller's offers as the catalog shows them: model, price, status and when each was checked.
  • POSTSeller Recheck/v1/provider/sellers/:key/recheckRead the seller's file again now instead of at the next daily run (at most once a day). Answers with the queued request.
  • GETSeller Analytics/v1/provider/sellers/:key/analyticsThe seller's own demand by day: views of its pages and offers, impressions in lists and clicks to its site, people and AI agents apart. JSON or CSV.
  • GETProvider/v1/providers/:provider_idOne provider: its card and every offer it sells.

account

  • GETMe/v1/meThe caller as the API sees it: the account (or anonymous), the key or token in use, its limits and what is left of them this hour and day.

meta

  • GETIndex/v1What /v1 offers (launch checklist 2026-09-30: llms.txt links here, so it answers, not 404): the endpoints, the OpenAPI description and the agents' guide.
  • GETMeta/v1/metaWhen the catalog was last collected, how many offers, models and providers it holds, and the license of the data.

API details

Base URL
https://airatecard.com
Authentication
An API key in the X-API-Key header
A bearer token in the Authorization header
OAuth 2.0 (authorization code flow) with 3 scopes
Operations
32 operations on 31 pathsGET 25POST 7
Main resources
provider 13catalog 9workload 4analytics 3meta 2account 1
Response formats
application/json
Description format
OpenAPI 3.1.0, API version 1

How to test the AI Rate Card in Swastik Codes

  1. Open the APISelect “Fork & run in Swastik Codes” on this page. Sign in or create a free account if asked.
  2. Fork it into your workspaceChoose a workspace. You get your own copy of the AI Rate Card collection, without the publisher's secrets.
  3. Add your credentialsThis API uses an API key in the X-API-Key header. Add yours in the collection's variables or an environment; the requests already reference it.
  4. Send a requestPick an endpoint, for example “Seller Analytics” (GET /v1/seller/:seller/analytics), and press Send. Check the response, add tests and run the whole collection with the runner.

Frequently asked questions

What is the AI Rate Card?

AI Rate Card (airatecard.com) is an independent catalog of AI model offers and prices, read daily from the providers' own pricing pages and files. Every… It is listed under AI & machine learning and has 32 endpoints.

Who publishes the AI Rate Card?

It is listed from AI Rate Card's public API description (an OpenAPI document), not published by AI Rate Card on Swastik Codes. Anyone who controls airatecard.com can claim the profile by adding a DNS TXT record.

What is the base URL of the AI Rate Card?

Requests go to https://airatecard.com. Endpoint paths are added to this base URL.

How do I authenticate with the AI Rate Card?

It accepts 3 methods: an API key in the X-API-Key header; A bearer token in the Authorization header; OAuth 2.0 (authorization code flow) with 3 scopes.

How many endpoints does the AI Rate Card have?

32 operations on 31 paths (25 GET, 7 POST). The largest groups are provider, catalog, workload, analytics.

Is forking the AI Rate Card free?

Yes. Forking copies the collection into your workspace for free. It has been starred 0 times and forked 0 times.

Suggest an API

Tell us which public API is missing. We check it, then add it here with its endpoints when it has an OpenAPI (Swagger) description.