惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

月光博客
月光博客
Microsoft Security Blog
Microsoft Security Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
博客园 - 叶小钗
博客园 - 三生石上(FineUI控件)
爱范儿
爱范儿
T
Tailwind CSS Blog
F
Fortinet All Blogs
Blog — PlanetScale
Blog — PlanetScale
WordPress大学
WordPress大学
H
Help Net Security
Apple Machine Learning Research
Apple Machine Learning Research
Y
Y Combinator Blog
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
GbyAI
GbyAI
Martin Fowler
Martin Fowler
有赞技术团队
有赞技术团队
腾讯CDC
V
V2EX
Jina AI
Jina AI
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
B
Blog
M
MIT News - Artificial intelligence
宝玉的分享
宝玉的分享

Latest from TechRadar

Quordle hints and answers for Monday, April 13 (game #1540) NYT Strands hints and answers for Monday, April 13 (game #771) NYT Connections hints and answers for Monday, April 13 (game #1037) Morbid Metal developer explains why he ditched an origami art direction in favor of gritty sci-fi — 'It worked, but it didn't really feel like me' '71% of US households get routers from ISPs': Why new FCC rules could leave millions stuck with outdated,… 'The CPU is the system’s executive layer': Intel joins SambaNova as both face existential threat from… ‘More bang for your buck’: 7 easy ways to boost your MacBook Neo’s performance for free DJI Romo P vs Roborock Saros 10R — which robot vacuum comes out on top when it comes to dodging obstacles? I put… I spent 6 hours with Genshin Impact on the Galaxy S26 Ultra, and I can't believe how far mobile gaming has come What is the release date for The Testaments episode 4 on Hulu and Disney+? I reviewed the LG G6 for 3 weeks, and it's a fantastic OLED TV that's the new best option for brighter rooms Is your bird feeder camera doing more harm than good? 3 tips for using it safely as RSPB issues urgent disease warning Chelsea vs Man City Live Streams: How to watch Premier League 2025/26 from anywhere in the world, team news How to watch Alcaraz vs Sinner for FREE: TV Channels for Monte-Carlo Masters Final Sunderland vs Tottenham Live Streams: How to watch Premier League 2025/26 from anywhere in the world, team news Are these the best-designed workout headphones ever? I used them for a month to find out How to watch Snooker 900 John Virgo online (it's free) – stream O'Sullivan vs Higgins anywhere I've only just discovered the Walk With Frodo app on Garmin's Connect IQ store — and as as a huge LOTR nerd, it's going to make the next 1,800 miles fly by 'Just not sustainable': Why your monthly £25 broadband internet bill could soon hit £45 How to watch Paris-Roubaix 2026: Free Streams & TV Info as Tadej Pogacar chases third Monument How to watch Euphoria season 3 online – stream Zendaya & Sydney Sweeney drama from anywhere today '$15K bill destroyed a solo developer’s startup': How hackers are using leaked Google API keys to… There's a sneaky way to watch UFC 327 really cheap... NYT Connections hints and answers for Sunday, April 12 (game #1036) NYT Strands hints and answers for Sunday, April 12 (game #770) Quordle hints and answers for Sunday, April 12 (game #1539) Amazon's Ring cameras are the perfect solution to secure your home on a budget — shop today's best deals… I've tested every iPhone since the iPhone 12, and Ceramic Shield 2 is the first iPhone glass I fully trust UFC 327 live stream: how to watch Procházka vs Ulberg, start time, preview, full card We're officially getting the DJI Pocket 4 on April 16, but here's how Insta360 could beat it
I pushed ChatGPT toward its hidden chat limit — here's wh...
Graham Barlow · 2026-06-03 · via Latest from TechRadar
Man looking at smartphone.
(Image credit: Getty Images / OMAR HAJ KADOUR )

You might not be aware, but your ChatGPT chat sessions have a hard limit. They can only go on for so long before you reach it. Confusingly there seem to be two types of limit as well. The first is to do with memory.

The memory limit (how much of the current chat that ChatGPT can remember for context) is not measured in time, or the number of messages you’ve typed, but in things called tokens.

A token is roughly 0.75 words in English, on average, so "The cat sat on the mat", which is six words, is about seven to nine tokens. Long responses, large tables, code, and detailed documents can consume far more tokens than short text exchanges.

What matters is the model's context window: this is the total number of tokens it can consider at once, and it will vary depending on the model you’re using.

The context window

OpenAI does not publicly document the context window for every model available in ChatGPT, but it is likely to be measured in hundreds of thousands of tokens. But if you reach it, then you can expect ChatGPT to start forgetting things from the start of your conversation.

Of course, if ChatGPT simply becoming forgetful about things way back at the start of the conversation was all that happened then there’d be no need to be alarmed. But as one Reddit user found out, there is another limit, that’s much more serious — ChatGPT can actually end a chat if it becomes too long.

When this happens you simply get a message saying, “You've reached the maximum length for this conversation, but you can keep talking by starting a new chat."

Sign up for breaking news, reviews, opinion, top tech deals, and more.

Despite there being no official documented chat length from OpenAI, there are a lot of reports and screenshots of this warning message, so while the evidence is anecdotal, it seems genuine and I think it’s still wise to act on it.

When you hit the limit

I’ve got one very long chat thread I use a lot in ChatGPT, pretty much every day, and I started to wonder when it was going to hit the limit, so I asked ChatGPT if the current chat was near the limits of its context window, and it said:

“I can't see the exact token count or the hard limit for this conversation, so I can't tell you "you're at 78%" or similar. That said, based on the amount of text in this thread, I'd estimate we're well past the halfway point and probably somewhere in the 60–80% range of what a typical long GPT-5.5 chat can comfortably hold before older context starts getting compressed or dropped.”

ChatGPT did estimate: “If I had to put a number on it, I'd say around 70% full, plus or minus quite a lot because OpenAI doesn't expose the actual count.”

So, it was probably time I started to think about opening a new chat based on the same subject.

How to prepare for a new chat

A chat thread can start to feel less like a sequence of prompts and more like a shared workspace that will always be there. In reality, it is still a temporary container with a maximum capacity. Eventually, the conversation gets too large, the context becomes harder to maintain, and the chat reaches the end of the road.

Basically, your only option to preserve the chat is to recreate it in a new chat window. That way your token usage resets to zero, and so does the conversation-length limit. Of course, you will lose a lot of the subtle bias or influence from the previous chat content, but it’s still your best option.

Because I’m getting near the limit of my chat, but not quite there yet, I’m at just the right point to do this now. The key thing is to do it early, before you hit your limit and it is too late to ask any more questions.

To recreate the chat, simply ask ChatGPT to create a prompt to recreate the current chat, then copy and paste what it produces into a new chat window at a time that’s convenient. Try this prompt: "Summarize everything important about this chat and create a starter prompt that recreates the context in a new conversation."

Now you can just start using the new chat.

If you have a ChatGPT conversation you've been using for months, it might be worth asking the AI to summarize it now rather than later. Because when the warning message finally appears, your only real option is to start over — and by then, the chance to recreate the exact chat may already be gone.


Google logo on a black background next to text reading 'Click to follow TechRadar'

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.


An Apple MacBook Air against a white background

Graham is the Senior Editor for AI at TechRadar. With over 25 years of experience in both online and print journalism, Graham has worked for various market-leading tech brands including Computeractive, PC Pro, iMore, MacFormat, Mac|Life, Maximum PC, and more. He specializes in reporting on everything to do with AI and has appeared on BBC TV shows like BBC One Breakfast and on Radio 4 commenting on the latest trends in tech. Graham has an honors degree in Computer Science and spends his spare time podcasting and blogging.