gpt-4.1 is not for everyone.

when you leave background, quality, and size blank—the image API inspects the user's prompt and intelligently picks the best params 🚢micro-ship: the response now echoes those choices, so you get smart defaults and full visibility

regularly blows my mind how easy it is to create images using up-to-date world knowledge Include `image_generation` and `web_search_preview` as tools in Responses then try a prompt like "Draw me a picture of downtown Seattle that shows the current time PDT and the weather"

We’ve made 13 price cuts since 2023 (same or better models) ✅ Plus batch + prompt caching ✅ 80% cuts? Just twice: - gpt-3.5-turbo - text-embedding-3-small Thrilled to see what you ship with today’s o3 drop

Imagegen high fidelity inputs are ideal for classic image editing tasks ✔️ Remove clutter, people, distractions 🔄 Swap objects 🎨 Change colors, textures, clothing 🌤️ Global tweaks: lighting, weather, time of day

High fidelity image editing is now in the OpenAI API and ChatGPT! Here's a quick demo I made that focuses on editing faces. It adds mustaches and facial expressions—but preserves all the other details of the face. Notice how my hair, glasses, nose, etc are perfectly maintained.

1/N I’m excited to share that our latest OpenAI experimental reasoning LLM has achieved a longstanding grand challenge in AI: gold medal-level performance on the world’s most prestigious math competition—the International Math Olympiad (IMO).

The team behind GPT-5 is much bigger than just OpenAI Dozens of partners worked with us to create evals + graders, hone prompts + ideal workflows, tune reasoning effort, verbosity, cost, latency—thank you 🤝 If you love GPT-5, that’s why. It takes a village.

this is the detail of GPT-5 I’m most proud of GPT-4 launched at $30/$60, no cache discount since then, it’s been an unrelenting cross-team push to collapse the cost of intelligence. we’re nowhere near done

Priority Processing debuts with GPT-5. under-hyped imo for apps where millisecond matters, pay extra and get our fastest token speeds just add "service_tier": "priority" to your requests

Good news! GPT-5 should now be significantly faster in Cursor. Big thank you to OpenAI for rolling out improvements to caching and API latency! P95 is ~2x faster versus before.