AIVideoFree Integrates OpenAI's GPT Image 2.6, Running Both New Models Alongside Competing Generators
AIVideoFree, a Singapore-based AI video and image generation platform, has integrated the latest version of OpenAI’s GPT Image model, specifically GPT Image 2.6, just one day after its release on November 4, 2026. This new integration brings both versions of the model, Sunburst and Flare, to AIVideoFree’s platform alongside other image models such as GPT Image 2.5 and Nano Banana 2.1.
AIVideoFree has made available two versions of the OpenAI model: Sunburst, which is optimized for output quality and maintains consistent details across repeated edits; and Flare, designed to operate at high speeds, capable of generating an image in approximately six seconds. This integration allows users to compare the performance of these new models with existing ones on AIVideoFree’s platform.
The GPT Image 2.6 model boasts a significant improvement over its predecessor, reportedly generating images about 50% faster. OpenAI has made this latest version available through their API in two versions: Sunburst and Flare. The speed at which the new models operate is particularly noteworthy, with Flare capable of producing an image in just six seconds.
AIVideoFree conducted a side-by-side text-rendering test to evaluate how well GPT Image 2.6 handles text within images, a known weakness for many image generators. A prompt was used that asked for a 3D-animated kitten in a Christmas setting with the French phrase ‘Merci pour ton inscription’ rendered as calligraphic lettering at the top of the image. The results showed how each model performed under identical conditions.
The test, conducted by AIVideoFree, aimed to provide users with practical insights into which models perform better and where improvements can be made. According to Roger, CEO of AIVideoFree, ‘Major image models are now being released every few weeks, and each one claims to be the best.’ He emphasized that running the same prompt through several models is a more effective way to assess their performance.
AIVideoFree plans to continue adding major image and video models as they become available. The platform also intends to publish comparison data showing how each model performs under identical prompts and settings, providing users with valuable insights into which models are best suited for specific tasks.
Discussion
No messages yet. Be the first to write.