Models & limits
Choose between Fable, Opus, Sonnet and Haiku, and learn how much fits in a chat, what Claude can't know, and when to start a new one.
-
What Can't Claude Know, and How Do You Fill the Gap? Claude's built-in knowledge stops at a cutoff, and Anthropic lists June 2026 as the reliable knowledge cutoff for Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5. Level 1: Basic prompts
- June 2026Reliable knowledge cutoff
- November 2026About five months laterthis stretch needs search or you
-
How Much Can You Fit in a Claude Chat? On paid plans, chats with Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5 have a 1M-token context window, and some older models have 500K or 200K. Level 2: Context
- 1System prompt, instructions, toolsconnectors take a lot
- 2Files and text you addedthe part you control
- 3Conversation so fargrows every turn
- 4Claude's reply and thinkingneeds room too
- Which Claude Model Should You Use? Claude comes in four models: Fable, Opus, Sonnet and Haiku. Level 1: Basic prompts
-
When Should You Start a New Claude Chat? Start a new chat when the topic changes, when Claude keeps returning to ideas you dropped, or when you're close to a usage limit. Level 1: Basic promptshow off track the chat is →New chat, handoff notename what you rejectedNew chatcarry nothing overKeep goingClaude can summarize older messagesNew chata short summary at mosthow far the topic has moved →
Three limits that are easy to confuse
Usage limits govern how much you can use Claude over time, with a five-hour session and a weekly limit on paid plans. The context window governs how long one conversation can get. The knowledge cutoff governs what Claude knows without searching, and for current models that's June 2026. Each has a different fix: wait or use less, start a new chat, or turn on search.
How to find out which one you hit
Check Settings > Usage first. If the session or weekly bar is full, it's a usage limit. If the message says the chat is too long, or Claude is "organizing its thoughts," that's the context window. If the answer is simply out of date, it's the cutoff. The cards in this lesson cover what to do in each case.
How long does Claude's limit take to reset?
On paid plans, the session limit resets five hours after your first message, and weekly limits reset once a week. Anthropic shows your remaining usage at Settings, Usage, and it may change limits at its discretion.
How big is Claude's context window?
On paid plans, chats with Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5 have 1M tokens, which Anthropic says is about 555,000 words. Some older models have 500K or 200K.
What is Claude's knowledge cutoff?
Anthropic lists June 2026 as the reliable knowledge cutoff for Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5. For anything later, turn on web search or paste the source.
Why does Claude say it's organizing its thoughts?
That message means compaction is running. As a chat nears the context limit, Claude summarizes earlier messages so it can continue. It needs code execution turned on, and long chats like this use more of your limit.