Factual recall success rates rose from 41.5% to 82.8%.
Jimmy Apples 🍎/acc#832
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
Positive users praise OpenAI's enhanced memory system for ChatGPT as a huge upgrade for better personal context and productivity across chats, while negative users object to forced buggy updates that disrupt legacy memories.
big upgrade to chatgpt memory rolling out today!
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
Better memory = Shorter prompts = More utility per token
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
much better ChatGPT memory:
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
Rolling out for both Pro and Plus users this morning. Real memory changes a lot of things.
'Today, we are launching a significantly more capable and compute-efficient memory architecture built on top of dreaming.
The memories synthesized by dreaming are reviewable through a summary of them made visible in the memory summary page. From the memory summary, you can quickly glean the highlights of what ChatGPT knows about you, add or update information about yourself, and provide instructions on what topics ChatGPT should bring up and when. If you want to drill down into a particular area to learn more, just chat with the model.'
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
Wait, this is actually a goated memory update, why isn’t anyone talking about this?
Claude has “Dreams” in its agent/API stack, but OpenAI is saying ChatGPT has had “dreaming” inside consumer memory since 2025.
And people might be confused on this update. Don’t worry I read the whole article:
Claude Dreams -> reorganize an agent memory store.
ChatGPT Dreaming -> synthesizes your personal context across chats, keep it fresh, and use it in normal conversations.
Now OpenAI is making Dreaming V3 the core memory layer for ChatGPT.
The benchmark jumps are crazy..
Recall: 41.5% → 82.8% Preferences: 31.4% → 71.3% Staying current: 9.4% → 75.1%
Here’s two real world examples of how the memory update looks
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
With the new memory system, you can review and steer what ChatGPT remembers through a memory summary, with more visibility and control over how context is used.
Your ChatGPT memories now benefit from a sleep wake cycle, which consolidates patterns and higher level context understanding to facilitate better answers!
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
We’re building ChatGPT to remember what matters, follow your preferences and constraints, and adapt as things change.
If you tell ChatGPT you’re planning a trip in July, memory should understand when the trip is upcoming, happening, and already over.
That helps ChatGPT keep giving relevant answers as context changes.
OpenAI gives ChatGPT a new dreaming memory system to retain preferences across conversations.
The big deal is that memory is becoming an active system, not a passive note file.
Earlier memory worked like a sticky note: it mostly remembered things you explicitly told it to remember, and those notes could become stale. If you said you were going to Singapore in July, the system might keep treating that as a current plan even after the trip ended.
Dreaming changes the model from “saving notes” to “maintaining a picture.” It looks across past conversations, extracts what still seems useful, updates what has changed, and makes that summary visible so users can review or correct it.
It measures factual recall: when a user asks something that depends on earlier chats, can ChatGPT correctly bring back the relevant personal context?
A better memory system means the assistant can start closer to where you actually are.
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
The new memory system will keep track of important details automatically. If you prefer the legacy saved memories experience, you can switch back in settings.
The new memory system is rolling out to Plus and Pro users in the US today, along with 2x more memory.
To access it on iOS or Android, update your ChatGPT app to the latest version. We’ll expand to more plans and countries soon.
@lostinlatencyX @sama @lostinlatencyX @yppif7 Hey! you should try tabula
You can keep ChatGPT’s internal memory off, but still bring in only the context you want when needed
Save, view, edit, or delete what’s remembered and carry the same memory across Claude, Grok, etc
http://www.tabula360.com
@OpenAI yawn
This is a major step.
Memory is not just personalization.
It is continuity.
And continuity is what turns AI from a session tool into a long-horizon collaborator.
But stronger memory also creates a new safety question:
does memory preserve the user’s trajectory,
or can it slowly reinforce drift?
The future of helpful AI will depend not only on remembering more,
but on keeping the human path understandable, stable and self-owned over time.
@gdb Yep, memory is good. I’ll give you that. But it’s pointless without 4o. May as well just give it us back instead of pushing out slop all the time, fix your company. #keep4o
https://openai.com/index/chatgpt-memory-dreaming/
@OpenAI
https://openai.com/index/chatgpt-memory-dreaming/
@OpenAI Memory is the next big moat after model cost. Let's see if this is good.
@ChrissGPT Memory is only useful if users can inspect and steer it. The moment it becomes invisible background behavior, every surprising answer turns into “what did the model think it remembered about me?”
@OpenAI OpenAI Korea B2C "ME"
Factual recall success rates rose from 41.5% to 82.8%.
Jimmy Apples 🍎/acc#832We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time.
Today, that work is rolling out as a more capable memory system in ChatGPT. https://openai.com/index/chatgpt-memory-dreaming/
