Open data: AI coding plan limits

Everything LimitWatch tracks, free to download as JSON or CSV: 26 plans (7 free tiers) across 10 vendors, 53 dated changes and 6 announced changes. Every row links to the vendor’s own source. Free to reuse under CC BY 4.0 with a link back.

Latest change 30 Oct 2026 · sources checked 2 Nov 2026 (source-check log) · no key, no sign-up, CORS enabled

Downloads

File	What’s in it
`plans.json`	Every tracked plan: price, every limit row, notes, vendor source and the date we last checked it.
`plans.csv`	The same plans, one row per plan (limit rows joined as “Label: value \| …”).
`plan-limits.csv`	One row per plan and limit (plan, label, value). Easiest for spreadsheets and pivot tables.
`changes.json`	Every dated, sourced change: before and after, direction, vendor wording, source, plans affected, and whether it was later reversed.
`changes.csv`	The same changes as CSV (plans separated by “;”).
`upcoming.json`	Changes vendors have announced with a date: scheduled, already in effect, or called off.

Base URL:

https://getlimitwatch.com/data/

License and credit

The data is licensed under Creative Commons Attribution 4.0 (CC BY 4.0). Use it anywhere, including commercially. Just credit LimitWatch with a visible link:

Credit line (HTML)

Data: <a href="https://getlimitwatch.com">LimitWatch</a>, CC BY 4.0

Credit line (Markdown)

Data: [LimitWatch](https://getlimitwatch.com), CC BY 4.0

Vendor quotes inside the data (the “vendor_framing” and “quote” fields) are short excerpts from the vendors’ own pages, kept for accuracy; they stay the vendors’ words. Plain text: “Data: LimitWatch (getlimitwatch.com), CC BY 4.0”.

Quick start

Shell (curl + jq)

curl -s https://getlimitwatch.com/data/plans.json | jq '.plans[] | {name, price, checked}'

Python (pandas)

import pandas as pd
limits = pd.read_csv("https://getlimitwatch.com/data/plan-limits.csv")
changes = pd.read_csv("https://getlimitwatch.com/data/changes.csv")
print(changes[changes.vendor == "anthropic"][["date", "title", "direction"]])

Fields

plans: slug, name, vendor, price, free, limits (label and value, as the vendor states them), notes, formerly (earlier names), source_url, checked (YYYY-MM-DD), url.
changes: id, date (when it took effect or was announced), vendor, plans, title, summary, before, after, direction (increase, decrease, neutral, unit-change, not-established, no-change), vendor_framing, source_url, via_url, verified (true: we loaded and checked the source page ourselves; false: carried from a secondary change ledger, cited as the source), later_update (e.g. reversed), url.
upcoming: date (first day the new terms apply), state (scheduled, in-effect, called-off), quote, source_url, checked. Also as a calendar: upcoming.ics.

Prefer updates as they happen? Use the RSS feeds, plan alerts or badges. Found a wrong number? Report it.

Open data FAQ

Can I use this data in my own site, app or article?

Yes. It is licensed under CC BY 4.0: you can copy, change and use it commercially, as long as you credit LimitWatch with a link to getlimitwatch.com.

How often is it updated?

The files are generated from the same data as the site, so they change as soon as a change is logged or a plan is re-checked. Vendor sources are checked daily; the last check was on 2 Nov 2026.

Where do the numbers come from?

From each vendor's own pricing pages, docs, help center and changelogs. Every plan and every change carries its source URL and the date we checked it. If a vendor publishes no figure, the row says so instead of guessing.

Is there an API key or rate limit?

No key and no sign-up. The files are cached at the edge and allow cross-origin requests, so you can fetch them from a browser. Please cache them on your side rather than fetching on every page view.