Hey everyone, we've got some exciting news! We've just upgraded the OpenAI models that power AtomAI.
For our premium users, you now have access to the brand new GPT-5 model. And for our free users, you're getting a big boost too, with an upgrade to GPT-5-mini.
So what does this actually mean for you? In short, AtomAI is now faster, more accurate, and more helpful across the board. No action or payment required, the new models are already active! Here’s a quick look at what you can expect.
What's New in Your Workflow
💬 Chat: More Accurate and Less Awkward
The Chat feature is now more reliable. The new model makes about 45% fewer factual mistakes (also known as hallucinations), so you can trust the answers you get a lot more. It also talks more like a person and less like a robot, and it's better at just saying "I don't know" instead of making things up.
📝 Write: Better Drafts, Less Editing
The Write feature also got a nice boost. It's now better at understanding what you want, whether you need a simple outline or help drafting a creative piece. Your generated drafts should have a better flow and sound more natural, which means less editing work for you.
📰 Read: Smarter PDF Summaries
Summarizing PDFs with the Read tool is much improved. The new AI is smarter at understanding documents that have charts and tables, not just plain text. This means you'll get more useful summaries and more accurate answers when you ask questions about your files.
🧠 Study: More Effective Study Tools
For the Study feature, this upgrade means better quality materials. The AI is now better at picking out the key concepts from your notes to create more relevant flashcards and practice questions. This should make your study sessions a bit more effective.
A Peek Under the Hood
For anyone curious about the tech specs, this new model is a pretty big deal. It's not just a small update. GPT-5 is now top-of-the-class on some really tough tests for AI.
- Expert-Level Reasoning: It's acing benchmarks that test reasoning on PhD-level science questions, showing it can handle very complex logic.
- Top Marks in Math & Coding: It has also set new records in advanced math and software engineering benchmarks, highlighting its powerful problem-solving abilities.
Bottom line: AtomAI is now faster, smarter, and more reliable for everyone. We're excited for you to give it a try and see how it feels. Log in and check out the new improvements!
You can read about the full list of changes on OpenAI's official blog post here.
