OpenAI [вернула]() пользователям ChatGPT оперативную память. Причём сразу несколько гигабайт. Раньше длинные локальные сессии в приложении ChatGPT могли съедать до 6 ГБ оперативки, причём потребление памяти продолжало расти вместе с перепиской. Теперь даже после 32 тысяч ответов в одной сессии приложение удерживает потребление памяти примерно на уровне 1,7 ГБ. Особенно полезно тем, кто запускает длительные сессии с ИИ-агентами на ноутбуках. 👍
Research1 min read
OpenAI has returned RAM to ChatGPT users. And several gigabytes at once.
OpenAI has returned RAM to ChatGPT users. And several gigabytes at once. Previously, long local sessions in the ChatGPT application could eat up to 6 GB of RAM, and memory consumption continued to grow along with correspondence. Now, even after 32 thousand responses in...
Telegram — Вайб-кодингBy: Вайб-кодингNovember 5, 2026 (1 hour ago)
Available in 4 languagesAI Summary & Key Takeaways
- OpenAI has returned RAM to ChatGPT users
- And several gigabytes at once
- Previously, long local sessions in the ChatGPT application could eat up to 6 GB of RAM, and memory consumption continued to grow along with correspondence
Research
KZ·AI 2870Инсайт дня
OpenAI has returned RAM to ChatGPT users. And several gigabytes at once.
Full text available in Russian
Tags:#OpenAI
Related stories
Local AI 301: Quantization
Learn about quantization in this livestream and get started with quantization using llama.cpp and vLLM. Program: - Introduction to Quantization, Merve Noyan, Hugging Face - Calibration based quantization, Colin Kealty (aka bartowski), RedHat - Quantization Aware Distillation,…
YouTube — Hugging Face
Inside Microsoft IQ: The Intelligence Platform Behind Agents - Copilot Dev Camp Summit Fall 2026
All right, thank you Paulo. Hi everyone. Welcome to our session about Microsoft IQ, the intelligence platform behind agents. My name is Pamela Fox. I'm a principal cloud advocate and I work across Azure Foundry, Copilot, and especially with uh Python applications. And I'm here…
YouTube — Microsoft Developer
GPT-6 Astra performed approximately 40% less work when programming tasks were presented as research...
GPT-6 Astra performed approximately 40% less work when programming tasks were presented as safety studies. This is stated in a new OpenAI report on deviations in the behavior of AI models. The researchers gave the model the same tasks, only changing them...
Telegram — Вайб-кодинг
Yann Lecun Was Right About LLMS.
For years, Yan Lakun has been telling the entire AI industry that it's making a huge mistake. While OpenAI, Google, Anthropic, and Meta spent hundreds of billions of dollars making large language models bigger and bigger, Lun kept saying the same thing. Bigger chat bots will not…
YouTube — The AI Grid