Skip to content

About AI Free API and our editorial policy

Who runs AI Free API, where the facts come from, how often they are re-checked, how the site earns money, and how to report an error.

What this site is

AI Free API (aifreeapi.com) is an independent third-party blog on how to use ChatGPT, Claude, Gemini and other AI tools, and how to fix them when something goes wrong. It launched in 2024 and publishes in six languages: Chinese, English, Russian, Japanese, Korean and Spanish. Articles answer the concrete questions people ask when they get stuck on OpenAI, Anthropic, Google and similar products: sign-in and verification, quotas and reset times, regional availability, what an error means and how to fix it, which plan to pick, and model IDs and pricing boundaries. Prices follow the providers' official pages; the site explains and compares, it is not the source of record.

The site is not affiliated with OpenAI, Anthropic, Google or any other provider. It does not run an API service, does not resell any provider's accounts, and does not recommend proxy-recharge, proxy-upgrade, shared or pooled subscription services. Services mentioned in articles are operated by the providers or partners named.

Where the facts come from

  • Prices, quotas, rate limits and model names follow the providers' official documentation: mainly the OpenAI developer docs, the Claude platform docs from Anthropic and Google AI for Developers, plus the official pricing pages and changelogs of image, video and speech models.
  • Every number gets a source next to it: the first time a price, limit or ranking appears, the article links to the page it came from, so readers can check it themselves.
  • No test we did not run is called a test: third-party benchmarks are credited with their source, settings and sample size; cost calculations show the formula and inputs so they can be reproduced.
  • Written by the AI Free API Team and checked against sources: each language is written on its own from one shared fact sheet rather than translated from another, so length and emphasis differ between languages. Before publication every number in the text is checked against the source it links to; claims that cannot be traced are cut or narrowed.

Updates and re-checks

Every article shows its publication date and its last update. When a provider changes a price, renames or retires a model, or moves a limit, the affected articles are revised; if a revision changes a conclusion, the article says so with a date. A dated notice inside an article means that section was checked on that day. Pages that get no traffic for a long time and are covered by another article are merged or removed, with the old URL redirecting (301) to the new page.

How the site earns money and how ads are marked

Revenue comes from promoting a partner service, LaoZhang API. The rules we apply:

  • Promotional blocks carry a visible "Sponsored" label, and every link to the partner's domains uses rel="sponsored".
  • Official prices and limits are written from the providers' documentation and are not changed because of the partnership.
  • Any service is recommended by the same evidence standard: what task it fits, which verified capabilities the recommendation rests on, and which conditions should make you choose something else.
  • We do not accept payment to change an article's conclusion.

The same team also runs the YingTu and LayerPSD websites.

What we do not do

We do not invent data or tests, do not copy other sites' articles, and do not publish content for getting around a provider's usage policies.

Corrections and contact

If a price, limit, model name or code sample is wrong, send us the article URL together with the official page you checked. We verify it, fix the text and move the "Updated" date; changes that affect a conclusion are noted in the article.

Email: hi@laozhang.ai