What does $20/mo actually get you? Daily-verified prices & usage limits for Claude Code, Cursor, Copilot, Codex + more. No guesses. [email protected]
Joined October 2026
- Tweets121
- Following13
- Followers1
- Likes23
New on LimitDex: Google Antigravity usage limits.
Free: weekly quota
AI Pro ($20/mo): 5-hour refresh, up to a weekly cap
Ultra $100: 5x Pro
Ultra $200: 20x Pro
Google doesn't publish exact quota sizes. Extra use runs on AI credits.
limitdex.com/antigravity-usa…
$100/mo, in vendor wording:
Claude Max: "five times the Pro plan's usage every five hours and every week"
Cursor Super: "15x more usage compared to Plus"
Copilot Max: "$200 per month" in AI credits
Codex Pro: "currently have no five-hour limit"
limitdex.com/ai-coding-plans…
OpenAI's help center: "Starting in December, Plus, Pro, and Business plans’ usage allowance will be shared across ChatGPT and Codex. Your usage for ChatGPT and Codex will be measured in tokens."
Instant chat won't count. Sizes aren't published yet.
limitdex.com/codex-usage-lim…
🔔 Plan change: Anthropic's pricing page no longer lists Max 20x.
Max, $100: "5x Pro usage every 5 hours and every week"
Ultra, $200: "No 5-hour limit" (verified accounts), "10x Pro usage every week"
Both include Fable "up to half your weekly limit".
claude.com/pricing
Cursor vs Codex, both start at $8:
Cursor Plus $8 (Grok models only)
ChatGPT Go $8 (no Codex limit stated)
$100: Cursor Super (15x Plus) vs ChatGPT Pro (no 5-hour limit)
Top: Cursor Ultra $200, ChatGPT Pro to $500
Neither publishes a fixed allowance.
limitdex.com/cursor-vs-codex…
Vibe coding on a budget? The app builders that do publish limits:
Bolt Free: 300K daily token limit, 1M tokens/month
Bolt Pro $25: 10M tokens/month, unused tokens roll over
Lovable Pro $25: 100 monthly credits
Every plan ≤$25:
limitdex.com/best-ai-coding-…
Is the $200 plan better value? Per dollar, using vendors' own multiples:
Claude Max 20x: 20x Pro's per-session allowance for 10x the price = 2x per $
Cursor Ultra: 50x Plus usage for 25x the price = 2x per $
Neither publishes the actual size.
limitdex.com/cursor-vs-claud…
5 of the 9 AI coding tools we track don't publish an exact usage limit:
Claude, Cursor, Codex, Grok, Devin.
They say things like "4x more usage." 4x of what?
We quote vendors verbatim and say "not published" when it isn't. Free JSON:
limitdex.com/data/plans.json…
LimitWatch #1 is out 📬
What your AI coding plan actually gets you: 9 tools, 37 plans.
A free weekly email on AI coding price and limit changes. No spam.
Subscribe: limitdex.com?utm_source=x&ut…
Cheapest paid way into each AI coding tool (monthly, checked today):
$8 Cursor Plus (Grok models only)
$8 ChatGPT Go (Codex; no limit stated)
$10 Copilot Pro ($15 credits)
$20 Claude Pro ($17 annual)
$20 Devin Pro
$20 Replit Core
$25 Lovable / Bolt
limitdex.com/best-ai-coding-…
How Claude Code limits work, per Anthropic's help center:
• Rolling 5-hour session window
• Weekly limit on top (Pro and Max)
• Max plans get a separate weekly limit for Fable
• Max 5x = 5x Pro's per-session allowance
Resets + what to do:
limitdex.com/claude-code-usa…
1/
There's no public "Opus 5.5 weekly limit" number. LimitDex reads Anthropic's pricing and help pages daily, and as of Oct 22, none of the pages we track gives a figure for Opus 5.5 or even names it. Here's what is published, and how to stay under a limit you can't see.
7/
Staying under it: 1) Know your reset day and do heavy Opus work early in the week. 2) Use /status to see what's left. 3) It's one pool, so claude.ai chats count too. 4) Use a lighter model for routine edits (our inference, not a vendor number).
8/
5) Measure real usage: npx ccusage@latest weekly --json. 6) At the wall: wait for the reset, turn on usage credits (Pro), buy API credits, or move to Max 5x. 7) Limits change without notice. Quotes, dates and sources: limitdex.com/claude-code-usa…