OpenAI launched Images 2.6 globally in ChatGPT and Codex, adding native GIF generation and reference-image support.

The reported rollout covers ChatGPT and Codex worldwide. OpenAI’s statements, relayed through Telegram, say Images 2.6 is available across subscription tiers on desktop, mobile, and web. The update adds reference-image support and native animated GIF generation.

The GIF workflow includes generation and editing within ChatGPT. Users can enter a text description or upload a photograph, then adjust playback speed, edit individual frames, and share the result. The report distinguishes this from earlier workflows in which ChatGPT helped create GIFs through code rather than a single native image-generation process.

Two developer models address different performance priorities. GPT-Image-2.6 Flare emphasizes speed, with a stated generation time as short as six seconds, while GPT-Image-2.6 Sunburst emphasizes image quality and consistency during editing. OpenAI also claimed 50% faster image generation, although the supplied message does not specify the benchmark conditions or comparison baseline.

The messages differ on application programming interface availability. One Financial_Express update says the two models are coming to the application programming interface, or API, while FinancialJuice explicitly reports that both are already accessible there. The supplied material contains no pricing, rate-limit, resolution, or compute-consumption details.