Skip to content

Repository files navigation

x-cmd/ratelimit — vendor API rate limits, referenced

A reference table of API rate limits and quota caps across the services the team uses most. This README is for visitors who land on the GitHub repo; the long-form articles and per-vendor deep-dives live under docs/ (served as the canonical site at x-cmd.com/ratelimit).

What this repo is

  • Quick reference for engineers. You came here to look up "Cloudflare's per-user API quota" or "GitHub's secondary rate-limit triggers". The data is under data/<vendor>.yaml; the per-vendor articles are under docs/.
  • Refreshable from upstream. Each YAML carries a verified flag and a docs_source URL. A future .github/workflows/scrape.yml job will re-pull official docs on a schedule, diff against the YAML, and open a PR when upstream numbers move.
  • Two-license. Source code, articles, scripts under Apache 2.0 (LICENSE). The rate-limit data tables under data/ under CC-BY-4.0 (LICENSE-data) — attribution required, commercial use OK.

Vendors covered

Vendor Surface covered Status Article
Cloudflare REST API + per-product HTTP caps unverified docs/1-cloudflare.en.md
阿里云 (Aliyun) OpenAPI per-product QPS unverified docs/2-aliyun.en.md
腾讯云 (Tencent Cloud) Cloud API 3.0 rate limits unverified docs/3-tencent.en.md
GitHub REST + GraphQL + Actions + secondary limits verified 2024-11 docs/4-github.en.md
Vercel Function/edge quotas + REST API unverified docs/5-vercel.en.md
BandwagonHost VPS-level port / bandwidth / connection caps unverified docs/6-bandwagonhost.en.md

At a glance

Vendor Primary API limit Window Headers
Cloudflare (REST) 1200 req 5 min / user Retry-After, cf-mitigated
Aliyun (open API) 100 QPS 1 sec / user custom Code field, not RFC 6585
Tencent Cloud (API 3.0) 20 QPS 1 sec / user X-RateLimit-*, Retry-After
GitHub REST (PAT) 5000 req 1 hr / token X-RateLimit-*, Retry-After
GitHub REST (unauth) 60 req 1 hr / IP X-RateLimit-*, Retry-After
GitHub GraphQL 5000 points 1 hr / token X-RateLimit-*, Retry-After
GitHub Search 30 req 1 min / user X-RateLimit-*
Vercel REST 1 req 1 sec / token lowercase ratelimit-* (RFC 9745 style)

⚠️ Numbers above are subject to verification. Each data/<vendor>.yaml carries a verified flag and the official-docs URL; the CI scraper will refresh them against upstream on every release.

Where to look next

  • data/<vendor>.yaml — machine-readable, the source of truth.
  • docs/0-ratelimit-overview.en.md — the canonical "how to use this repo" article.
  • docs/<n>-<vendor>.md — per-vendor deep dive.
  • RATELIMIT-RESEARCH.md — working notes, including what the team has verified vs. what's still pending.

License

  • LICENSE — Apache 2.0 (code, prose, scripts).
  • LICENSE-data — CC-BY-4.0 (data files under data/).

Contributing

See CONTRIBUTING.md. External PRs are welcome for data corrections, new vendor files, and article edits; the team signs off on every change.

About

Reference I built for figuring out API rate limits while designing integration plans — Cloudflare, Aliyun, Tencent, GitHub, Vercel. https://x-cmd.com/ratelimit

Topics

Resources

Contributing

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors