ChatGPT Is Getting a GIF Maker Powered by OpenAI's Unannounced Images 2.6 Model

What it claims

Decrypt, citing The Verge, reports that OpenAI is gradually rolling out a GIF generator inside ChatGPT, powered by Images 2.6. OpenAI has not formally announced the model. The Verge reports Images 2.6 is coming to ChatGPT and Codex users on web, desktop and mobile, and that the broader update promises faster image generation. For developers, two API models are planned: GPT-Image-2.6 Flare, aimed at speed, and GPT-Image-2.6 Sunburst, aimed at quality. The page gives no pricing, rate limits, API parameters, output formats or release dates for either model.

Evidence

The page has no benchmarks for Images 2.6. Decrypt could not get the GIF feature on its accounts. In its test, the model produced a series of frames but no assembled animation.

The only measured figures are for the predecessor, Images 2.5, released in September. OpenAI claimed sharper details, more precise editing and generation up to 50% faster than the model before it. In Decrypt's earlier six-category comparison against Google's Nano Banana 2, each model won three categories. Decrypt credits Images 2.5 with strong illustration quality and less oversharpening, but says it still made mistakes rendering text. Nothing on the page says whether 2.6 fixes the text rendering problem.

Limits

The report is secondhand: the main source is The Verge, and no OpenAI announcement or documentation is quoted. No capability, speed or quality claim for Images 2.6 has been tested independently. OpenAI has given no date for when all ChatGPT accounts will get GIF generation. The page notes that GIFs are much simpler than the video output of Sora, the standalone app OpenAI shut down in March, so the feature is not a replacement for video generation.