The AI limits tracker

AI plan usage limits, tracked and dated

The actual message caps, weekly caps, and context windows behind every major AI subscription — each fact individually verified, dated, and linked to its official source. When a limit changes, the changelog records it.

65 tracked factsPer-fact verification datesLast verified Jul 29, 2026

Popular paid plans: the limits that matter

Compact view of the most-compared paid tiers. The full comparison covers all 16 tracked plans, free tiers included.

Full limits table
PlanPriceMessages / requestsWeekly capContext window
ChatGPT Plus
OpenAI
$20/month160 messages per 3 hoursUp to 3,000 per week32K tokens
ChatGPT Pro
OpenAI
$200/monthEffectively unmetered128K tokens
Claude Pro
Anthropic
$20/month~45 messages per 5 hoursApplies across all models200K tokens
Claude Max
Anthropic
$100-$200/monthMessages (Max 5x, $100/mo): ~225 messages per 5 hoursMessages (Max 20x, $200/mo): ~900 messages per 5 hoursWeekly usage cap (Max 5x): 5× Pro's weekly allowanceWeekly usage cap (Max 20x): 20× Pro's weekly allowance200K tokens
Google AI Pro / Gemini Advanced
Google AI Pro
$19.99/month4× the standard allowance1M tokens
Perplexity Pro
Perplexity
$20/month300+ per day

✓ Figures verified between Jan 15, 2026 and Jul 29, 2026 against official provider pages · hover any value for its date · full per-fact dates and sources in the plan sections

Recent limit changes

Providers change limits quietly. We write them down.

Full changelog
  1. All plans

    Full verification pass against live provider pages: ChatGPT Plus 160/3h and 3,000 Thinking/week reconfirmed; Claude Pro's ~45 messages/5h and all-model weekly cap reconfirmed; Gemini and Copilot facts updated for their 2026 metering changes below.

  2. Copilot Pro

    GitHub replaced Copilot's monthly premium requests with GitHub AI Credits: Pro now includes 1,500 credits (≈$15 of model usage) per month, metered by token consumption at each model's rate.

    Before: 300 premium requests / month ($0.04 per extra). After: 1,500 AI Credits / month (1 credit = $0.01).

    Source
  3. Gemini Free

    In May 2026 Google replaced Gemini's fixed daily prompt counts (5/day free, 100/day AI Pro, 500/day Ultra) with compute-based limits that refresh every 5 hours inside a weekly cap; AI Pro is now '4× standard' and Ultra '5×–20× Pro'.

    Before: Fixed daily prompt counts per tier. After: Compute-based allowance, 5-hour refresh + weekly cap.

    Source

Not sure which plan fits? Use the finder

Answer three questions about your workflow and budget, then check the recommendation against its tracked limits.

I mainly need AI for…
My budget
What matters most (pick any)
Best match

ChatGPT Plus

$20/month

Flagship-model messages: 160 messages per 3 hours

ChatGPT Plus is the strongest fit among standard paid plans for a general workflow. Your selected needs point toward writing quality.

Limits by provider

Tier-by-tier breakdowns with full fact tables and sources.

OpenAI

ChatGPT usage limits

~10 messages/5h free, 160/3h on Plus, effectively unmetered on Pro — plus weekly reasoning caps and context tiers.

ChatGPT limits
Anthropic

Claude usage limits

~45 messages/5h on Pro, 5x/20x multipliers on Max, weekly caps since August 2025 — and what happens when the window runs out.

Claude limits
Google

Gemini usage limits

Compute-based since May 2026: allowances refresh every 5 hours inside a weekly cap — AI Pro gets 4× standard, Ultra 5×–20× Pro.

Gemini limits

FAQ

Concrete answers from the tracked data.

Which AI plan has the highest usage limits for $20/month?

By message volume, ChatGPT Plus (160 messages per 3 hours) beats Claude Pro (~45 per 5 hours). By context depth, Claude Pro's 200K tokens beats ChatGPT Plus's 32K. Perplexity Pro allows 300+ Pro searches per day, and Google AI Pro gives 4× Gemini's standard compute allowance. The right answer depends on which limit your workflow actually hits.

How is this data verified?

Every fact carries its own verification date and a link to the official provider page it came from. Where a provider publishes no number, the table says 'Not published' instead of guessing. Changes are recorded in the changelog with before/after values.

Do AI usage limits actually change often?

Yes, constantly. In 2026 alone, Google scrapped Gemini's daily prompt counts for compute-based limits (May) and GitHub replaced Copilot's premium requests with AI Credits (June). In 2025, Anthropic added weekly caps to Claude and OpenAI raised the Plus reasoning cap from 200 to 3,000 messages per week. That churn is exactly why each fact is dated.

Machine-readable

Use this data in your own tools

The full limits dataset and changelog are available as JSON — free to use with attribution. Updated Jul 29, 2026.