Halfway through a very long analysis Claude says you have reached a limit, yet a brand-new chat you open straight afterwards works normally. What did you actually run into?

The principle — Claude has two limits that produce similar-sounding messages and call for opposite responses.

Why the key is correct — Anthropic draws the distinction plainly: usage limits control how much you can use Claude across all your conversations, while length limits control how long any single conversation can become. A new chat opening normally is the diagnostic. If your usage allowance were spent, nothing would work; because the fresh chat is fine, what you filled was one conversation's context window rather than your allowance.

Why the others are wrong — a usage limit does not propagate slowly, nothing caps how many chats you may open, and retrying inside the same conversation resends the very thing that overflowed.

Remember this — the responses differ, so the diagnosis matters. A usage limit means waiting, upgrading or buying credits. A length limit means starting a new conversation, or moving the material into a project so it does not have to sit in the chat at all.

Sources — Anthropic's article on how usage and length limits work.