OpenAI adds a file Library to ChatGPT for storing, searching and reusing files
OpenAI has launched a Library tab in ChatGPT that stores every uploaded and generated file, from PDFs and spreadsheets to presentations and images. You can reattach files to new conversations, search them by type and origin, and download or delete them separately from chats. The full Library is currently available only on the web, for Plus, Pro and Business subscribers.
AI-processed from ZDNet AI; edited by Hamidun News
OpenAI has added a dedicated file library, Library, to ChatGPT. Documents, spreadsheets, presentations and other attachments are no longer tied permanently to a single conversation: you can find them later, attach them to a new chat and use them as persistent working material.
How Library works
The rollout began on March 23, 2026. The logic is simple: every file a user uploads to a regular chat, or that ChatGPT creates on request, is saved to Library automatically. On the web, the section sits in the left sidebar, where you can open a list of all files, pick a document, download it or delete it without touching the original conversation.
The practical value shows up in new conversations. Through the attachment button you can now choose recent files or open Add from Library to quickly pull in an old file without scrolling through dozens of chats. ChatGPT can also answer questions about previously saved materials, such as "what's important in the document I uploaded yesterday."
On iOS and Android, file search and the list of recent attachments already work, but the full Library interface is so far available only on the web.
What goes into the library
Library holds more than PDFs and text documents. It also collects spreadsheets, presentations, images and other files uploaded to standard ChatGPT conversations. If the service itself created a file in a response, for example by exporting a spreadsheet, building a presentation or generating a document, that artifact is saved too. The exception is generated images: they still live in the separate Images tab rather than in the general library.
The new tab is not just a warehouse. It offers basic tools for navigation and control:
- filtering by uploaded and generated files
- filtering by type: images, documents, spreadsheets, presentations and PDFs
- a limit of up to 512 MB per file
- a limit of about 50 MB for CSVs and spreadsheets
- a limit of 20 MB per image
The feature is currently rolling out to Plus, Pro and Business subscribers and is not yet available in the European Economic Area, Switzerland or the United Kingdom. For everyone else, it looks like a notable step from one-off chats toward a persistent workspace: upload a set of project materials once, then return to them in different conversations instead of starting from scratch.
What to know about storage
The most notable change concerns not the interface but what happens to data after chats are deleted. Many people used to treat an attachment as part of a specific conversation: delete the thread and you lose the file. The rule is now different. A file stays in the account until the user deletes it manually from Library. Routine history cleanup no longer wipes the file archive.
After a file is deleted from the library, it disappears from the account immediately, and permanent deletion from the company's systems should normally occur within 30 days, unless there are legal obligations or security-related reasons to keep it longer.
There are exceptions: attachments from Temporary Chat do not go into Library, and where ChatGPT Health is available, files from that mode are not saved either.
Data settings are a separate nuance. Saved files and their associated chats are subject to the same data controls as the rest of the user's content. If the Improve the model for everyone option is on, submitted materials may be used to improve models. For teams and professionals who run work documents through ChatGPT, this is no reason to skip the feature, but it is a reason to check account settings and internal rules on confidential data right away.
What this means
Library looks like a minor interface update, but it is an important shift. ChatGPT looks less and less like a chat that starts from a blank screen every time, and more like a workspace where files, results and context accumulate. For users who analyze documents, compile reports and return to the same materials for weeks, it removes needless re-uploads and makes ChatGPT noticeably more useful in long work scenarios.
- Nov 1, 2026Local models lose to Opus and Codex: 0.21 on a MacBook versus 1.00 for Codex
- Nov 1, 2026Kimi scores 96.2% on a 1C database analytics test and emerges as a fallback if Claude access is blocked
- Nov 1, 2026OpenAI Rewrites GPT Behavior Rules: Five Values and a Chain of Command
- Nov 1, 2026Google admits AI Mode cuts search revenue, shares fall 15%
Key Questions Answered
How Library works?
The rollout began on March 23, 2026. The logic is simple: every file a user uploads to a regular chat, or that ChatGPT creates on request, is saved to Library automatically. On the web, the section sits in the left sidebar, where you can open a list of all files, pick a document, download it or delete it without touching the original conversation. The practical value shows up in new conversations. Through the attachment button you can now choose recent files or open Add from Library to quickly pull in an old file without scrolling through dozens of chats. ChatGPT can also a
What goes into the library?
Library holds more than PDFs and text documents. It also collects spreadsheets, presentations, images and other files uploaded to standard ChatGPT conversations. If the service itself created a file in a response, for example by exporting a spreadsheet, building a presentation or generating a document, that artifact is saved too. The exception is generated images: they still live in the separate Images tab rather than in the general library. The new tab is not just a warehouse. It offers basic tools for navigation and control: filtering by uploaded and generated files filte
What to know about storage?
The most notable change concerns not the interface but what happens to data after chats are deleted. Many people used to treat an attachment as part of a specific conversation: delete the thread and you lose the file. The rule is now different. A file stays in the account until the user deletes it manually from Library. Routine history cleanup no longer wipes the file archive. After a file is deleted from the library, it disappears from the account immediately, and permanent deletion from the company's systems should normally occur within 30 days, unless there are legal obl
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.