The platform added the Sunburst and Flare models and ran a side-by-side text-rendering test against other image models.
-- AIVideoFree, an AI video and image generation platform based in Singapore, has integrated GPT Image 2.6, the image generation model OpenAI released on November 4, 2026. Both versions of the model, Sunburst and Flare, are now available on the platform next to other image models, including GPT Image 2.5 and Nano Banana 2.1.
About the new model
OpenAI says GPT Image 2.6 generates images about 50% faster than its predecessor. The model is offered through OpenAI’s API in two versions. Sunburst is tuned for output quality and is designed to keep details consistent across repeated edits. Flare is tuned for speed and, according to OpenAI, can return an image in about six seconds.
On the day of release, DesignArena reported that Sunburst had moved to first place on its Image Arena leaderboard, ahead of the previous GPT Image 2.5 Sunburst.
Side-by-side test
To see how the new model handles text inside images, a known weak point for image generators, AIVideoFree ran the same prompt through both GPT Image 2.6 versions and other models on its platform. The prompt 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.
[Attach comparison images: GPT Image 2.6 Flare／MAl Image 2.6／MAl Image 2.6 Flash／Grok Imagine]
Full prompt: An adorable Norwegian kitten is sitting across from each other. The kitten displays a cheerful expression in a Christmas setting. The style is a 3D animation of exceptional quality, with fluid rendering and ultra-detailed fur textures. At the top of the image, an elegant and stylized calligraphic lettering with a subtle effect in relief displays the French text "Merci pour ton inscription". The composition is centered, highlighting the cat with an irresistible "kawaii" aesthetic. Christmas sweetness typical of high-end illustrations from modern animation studios.
“Major image models are now being released every few weeks, and each one claims to be the best,” said Roger, CEO of AIVideoFree. “Running the same prompt through several of them is the most practical way to see where a new model actually improves, and that is the approach we plan to keep taking with each release.”
Next steps
AIVideoFree said it plans to continue adding major image and video models as they are released and to publish comparison data showing how each model performs under the same prompts and settings.
About AIVideoFree
AIVideoFree is an online AI video and image generation platform based in Singapore. It offers access to multiple video and image models, including Veo, Kling, Seedance and GPT Image, from one interface. New users receive free credits that work on every model, and outputs carry no watermark. Website: https://www.aivideofree.com
Media contact
Roger, CEO, AIVideoFree
Singapore
GPT Image is a trademark of OpenAI. AIVideoFree uses OpenAI’s API and is not affiliated with or endorsed by OpenAI. Model performance figures are as reported by OpenAI and DesignArena.
Contact Info:
Name: Roger lee
Email: Send Email
Organization: AIVideoFree
Website: https://www.aivideofree.com
Release ID: 89208567
In the event of encountering any errors, concerns, or inconsistencies within the content shared in this press release, we kindly request that you immediately contact us at [email protected] (it is important to note that this email is the authorized channel for such matters, sending multiple emails to multiple addresses does not necessarily help expedite your request). Our dedicated team will be readily accessible to address your feedback within 8 hours and take appropriate measures to rectify any identified issues or facilitate press release takedowns. Ensuring accuracy and reliability are central to our commitment.
