How much Claude Code do you get on the Pro plan?

Nobody can give you a reliable number of hours or messages, because Anthropic does not publish one. What Anthropic does say is that Pro has a session limit that resets every five hours plus a weekly limit, that Pro gives "at least 5x more usage per 5-hour session than Free", and that Claude Code shares those limits with the rest of Claude. How far that goes depends on your work. This is accurate as of November 2026.

What Anthropic publishes

Per Anthropic's pricing page, Pro costs $20 a month ($17 a month billed annually, US prices), and its FAQ says Pro gives "at least 5x more usage per 5-hour session than Free". Per the usage and length limits article:

• A session limit resets every five hours. The period starts when you send your first message, and the limit resets in full five hours later.
• Weekly limits reset once a week, at a fixed time assigned to your account.
• Usage across claude.ai, Claude Code and Claude Desktop counts toward the same limit.

Beyond that, Anthropic publishes nothing about the size of Pro's limits. There is no official table of messages per session, tokens per week or hours of Claude Code. If you see one on a blog or forum, ask where it came from.

Why the old hour estimates are outdated

In July 2025, when weekly limits were announced, Anthropic estimated that "most Pro users can expect 40 to 80 hours of Sonnet 4 through Claude Code" a week, as reported by TechCrunch. Those figures still circulate. They should not be used now, because:

• The models changed. The estimate was for Sonnet 4. As of November 2026, Claude Code's default model on Pro is Opus 5.5, and Anthropic says Opus "uses meaningfully more of your quota" than Sonnet.
• Limits changed. On May 6, 2026 Anthropic doubled Claude Code's five-hour limits for Pro, Max, Team and seat-based Enterprise plans. Claude Code's weekly limits were 50% higher from May 13 through September 13, 2026, and from September 14 they are 25% higher than before the promotion, per Anthropic's notice.
• Anthropic says it may adjust limits at its discretion to manage capacity.

Anthropic has published no current hour figures. This post does not attempt to replace them with a guess.

What changes how fast you use it

Anthropic lists several factors in the usage article: the length and complexity of your conversations, the features you use, which model you are using, and the effort level. The best-practices articles add file attachments, tool usage and multi-step tasks. For Claude Code specifically, Anthropic's cost guide explains why a long session gets expensive: the long context is resent with every request, and after a break longer than the cache lifetime the next request misses the cache.

The practical levers, all from Anthropic's docs:

1. Model. "Opus costs several times more per turn than Sonnet, and Sonnet more than Haiku." Spending Opus on routine work is the fastest way to drain a limit. /model changes it, and opusplan plans with Opus and executes with Sonnet.
2. Effort. "Higher effort uses more tokens." Change it with /effort.
3. Context. Use /clear between tasks. Anthropic's rule of thumb: use /clear when starting a new task, and /compact when continuing a long one.
4. Tools. The Help Center says "Tools and connectors are token-intensive". In Claude Code, /context shows what is taking up space and /mcp lets you disable MCP servers you do not use.
5. Delegation. Subagents still draw on your usage, and agent teams use about 7x more tokens than standard sessions when teammates run in plan mode.

More in how to make Claude Code limits last longer.

Fable and other special cases

Fable models are not included in Pro's plan limits. They run on pay-as-you-go usage credits from the start, apart from the Fable 5.2 launch week that ends on November 16, 2026. See can you use Fable on the Pro plan. Also, claude -p and the Agent SDK still draw from your plan's limits when signed in with the plan. If an ANTHROPIC_API_KEY is set in your environment, Claude Code may bill API usage instead of using Pro at all.

How to measure your own

Because there is no published figure, the honest answer is to measure.

1. In Claude Code, run /usage. It shows plan usage bars for the session and the week, plus a breakdown of what counts against your limits and recommendations for behaviours behind 10% or more of recent usage.
2. For an account-wide view, open Settings > Usage on claude.ai, which shows "Current session" with percent used and time remaining, and the weekly limits.
3. Note the percentage before and after a typical task. A few days of that tells you more than any estimate.

The /usage breakdown is computed from local session history on the machine you run it on, so use from other devices or claude.ai is not included in the breakdown, although the plan limit bars are account-wide.

What to do when you run out

When you hit a limit, Claude Code blocks further requests until the reset time in the message. Anthropic's options for Pro are: consider Max, turn on usage credits, switch to a Console account and buy API credits, or wait. See Pro vs Max for whether the upgrade is worth it, and what 5x Pro usage means for how the multipliers on bigger plans are measured.

Users have also reported limits being reached sooner than expected or usage climbing while idle, for example in GitHub issues #61012 and #108661. These are reports from individual users, not documented behaviour, but they are another reason to measure your own usage.

Keeping an eye on the number

Headroom is a free Mac menu bar app that shows the session and weekly percentage used for each Claude Code account signed in on your Mac, with the reset countdown, so you can note the before and after of a task without running a command. It never sends a prompt, so checking does not use your limits. It is independent, from Everyday Mac Tools, and not made or endorsed by Anthropic. Get it from the Headroom home page.