Does Claude remember your past conversations?

It can, in two different ways. Memory lets Claude save what it learns about you as you chat and carry it into new conversations. Chat search lets Claude look through earlier chats when you ask about them. Here's who gets each, where the switches are, how to check or fix what Claude knows, and where a written context file still helps, based on Anthropic's own help pages.

Anthropic recently moved everyone to a new memory experience. If your settings show Memory as its own section under Settings, the steps below apply. A few Team and Enterprise organizations still have the older version, which lives under Settings, Capabilities. Menus change; treat setting names as signposts.

Memory vs chat search

• Memory is on by default for Free, Pro and Max on the web, Claude Desktop and Claude Mobile. On Team and Enterprise plans, an owner decides whether it's available, and it stays off for each member until they turn it on.
• Chat search ("Search and reference chats") is for paid plans only: Pro, Max, Team and Enterprise, on the web, desktop and mobile apps. It's on by default once it reaches your account.

So on the Free plan, Claude can build memory from your chats but can't search back through old conversations for you.

How Claude's memory works

Claude saves memory as individual topics while you chat, rather than summarizing a conversation after it ends. Anthropic's example: mention that a deadline moved, and your next conversation already knows. Claude saves on its own, and you can also say "remember this" to save something directly.

According to Anthropic, Claude remembers everyday context such as:

• Your role, projects and professional context
• The people and places in your work and life
• Communication preferences and working style
• Technical preferences and coding style
• Project details and ongoing work

By default it does not store sensitive topics like health, religious beliefs or politics. You can opt in with Include sensitive topics in memory in Settings, Memory. Some things are never saved even if you ask, including government ID numbers, financial account numbers, criminal history and immigration status.

Each Claude Project keeps its own memory. Inside a project, Claude uses that project's memory plus your memory from chats outside projects, but not memory from other projects.

Turn memory on or off

1. Open Settings, then Memory.
2. Switch on Generate memory from chats to turn it on.
3. To turn it off, click the same switch and pick one of two options:
   • Pause memory: Claude keeps what it already has but stops using it or adding to it. Chats while paused won't be added later.
   • Reset memory: permanently deletes all memories, including project memories. This can't be undone.

Ask Claude about a past chat

On a paid plan, just ask in a new chat. Anthropic's own examples:

What did we discuss about [topic]?
Can you find our conversation about [subject]?
Let's continue where we left off with [project].

When Claude searches, it shows up in the chat as a tool call, with citations that link back to the original chats. Outside projects, Claude searches all your chats outside projects. Inside a project, it only searches that project's chats. To stop searches, turn off Search and reference chats in Settings, Memory.

See, edit or delete what Claude remembers

• In settings: go to Settings, Memory. Everything Claude remembers is listed under Topics. Open a topic, then use the edit icon to change it or Delete to remove it. A fix applies to every conversation from then on.
• In a chat: tell Claude what to remember, change or forget. The update applies to your next conversation.
• Deleting a chat isn't enough. When a conversation is deleted or expires, memory entries made from it are not removed. Delete those memories yourself in Settings, Memory.

Keep one chat out of memory

You have two options, and they behave differently:

• Turn off Memory for one chat: start a new chat, click + in the message box, and switch off Memory before you send the first message. Claude won't use or add to memory there, but the chat stays in your history and can still turn up in searches from other chats. It also works inside projects.
• Incognito chat: outside a project, click the ghost icon in the upper right of a new chat. It isn't saved to your chat history or memory, and Claude won't find it in later searches. It's available on every plan. Anthropic says incognito chats are still kept for 30 days by default, and once closed they can't be reopened.

Memory vs Instructions for Claude

Memory is written by Claude, from what comes up in your chats. Instructions for Claude are written by you, in Settings, and apply to every chat. Anthropic's incognito help page notes that Claude can still read your profile preferences inside an incognito chat, even though it skips memory there. For things Claude must always know, like your role and how you like answers, put them in Instructions for Claude. See the Claude setup guide for where that lives.

The fix that works everywhere

A personal context file is a short document about you that you write once and control. Put the short version in Instructions for Claude, upload the full version to any project, and reuse the same file in ChatGPT, Gemini or Grok. In any new chat, you can paste this line with your file underneath:

Here is my personal context file. Read it and use it for everything in this chat. You don't need to summarize it back to me. Just reply "Got it" and wait for my first request.

[paste your context file here]

Sources: Anthropic Claude Help Center, Use Claude's chat search and memory to build on previous context and Use incognito chats.