OpenAI Releases Images 2.6 With GIF Output And 50 Percent Faster Generation
OpenAI announced Images 2.6 on November 4, 2026, and the image generation model is rolling out to ChatGPT and Codex on web, desktop, and mobile, including the free tier, according to the announcement.
2 sources 1h ago
Reporting from 2 sources: ASCII.jp, GIGAZINE.
Why it matters
Image generation was already a weekly habit at scale, with OpenAI citing more than 4 billion images created per week across ChatGPT, Codex, and the API, and Images 2.6 prices API tokens the same as Images 2.5.
Key Facts
- OpenAI announced Images 2.6 on November 4, 2026, and it rolls out to ChatGPT and Codex on web, desktop, and mobile across all subscription tiers including free.
- OpenAI says Images 2.6 generates images about 50 percent faster than the previous Images 2.5 model.
- The API offers two models, GPT-Image-2.6 Flare and GPT-Image-2.6 Sunburst, with Flare able to generate an image in as little as 6 seconds at low quality.
- Images 2.6 adds GIF animation generation and a complexity parameter that adjusts generation speed, image quality, and processing complexity.
- OpenAI states that more than 4 billion images per week and more than 500 million images per day are created through ChatGPT, Codex, and the API.
OpenAI says Images 2.6 generates images about 50 percent faster than the prior Images 2.5 model. The release adds GIF animation generation, so a simple sketch can become an animated file, and a new complexity parameter that lets users trade generation speed, image quality, and processing complexity against each other. The API ships two models: GPT-Image-2.6 Flare, a lightweight edition that can produce an image in as little as 6 seconds at low quality, and GPT-Image-2.6 Sunburst, intended for work where detail and repeated editing need to stay consistent. Image markup, already present in ChatGPT, arrives in the API so users can flag the exact spots to correct. Output can also be SVG.
Both model variants ship through the same API at a price OpenAI kept level with the previous generation: 1 million tokens on Image-2.6 costs the same as on Image-2.5. The company puts current usage at more than 4 billion images a week and more than 500 million a day across ChatGPT, Codex, and the API.
The functionality splits by channel rather than by plan. The free tier gets the full model, and the differences appear in what each surface exposes:
- ChatGPT and Codex: GIF generation from a simple sketch, plus the complexity setting that balances speed, quality, and how much composition the model weighs.
- API, GPT-Image-2.6 Flare: the lightweight option for quick prototyping, reaching an image in as little as 6 seconds at low quality.
- API, GPT-Image-2.6 Sunburst: built for detail rendering and repeated edits, where consistency across passes matters more than raw speed.
- API, image markup: the marking tool already in ChatGPT moves over, letting users flag the spots to correct instead of describing them in text.
- SVG output: vector files whose outlines hold up when enlarged, avoiding the degradation a raster image shows at scale.
OpenAI frames the update as finer control across the whole pipeline, from creating an image through editing it to handing a project off to the next stage.
Written with AI by Yomimono from the 2 cited sources below, including Japanese-language reporting where cited. See our editorial policy.
Add Yomimono as a preferred source on Google