xAI's Aurora Image 2.0 debuts at #2 on Text-to-Image Arena leaderboard with 1,320 points
The model's low-quality setting outperforms the previous generation's best setting across multiple benchmark categories on launch day. API access is absent, limiting use to xAI's own app.
xAI launched Aurora Image 2.0 on August 8, 2026, featuring precision editing, crisp text rendering, and improved factuality for real-world use. The model's low-quality setting debuted at #2 on the Text-to-Image Arena leaderboard with 1,320 points, a jump of 12 places over the previous Grok generation. The older Grok Imagine Image Quality model dropped to #14 with 1,228 points, meaning the new model's low setting outperforms the prior generation's careful setting by 92 points. Aurora Image 2.0 also debuted at #2 in the Image Edit Arena with 1,439 points, ranking #2 across Art, Portraits, Text Rendering, Cartoon/Anime/Fantasy, and Photorealistic and Cinematic Imagery sub-categories, and #3 in Product, Branding and Commercial Design. Access is limited to xAI's app and is not available via API.
Microsoft's MAI-Image-2.6 entered the Text-to-Image Arena on August 10 at #2 with 1,336 points, sitting 45 points behind GPT Image 2 at #1 and pushing xAI's Grok Image 2.0, which had debuted at #2 on August 8, down to #3.
Microsoft plans Playground availability and early API access on Foundry. Alibaba's Qwen-Image-3.0-Pro had also entered the competitive period, reaching #5 globally on August 4 and later ranking #1 among Chinese models and #2 among mainstream models, according to Alibaba.
MAI-Image-2.6 enters Text-to-Image Arena at #2 with 1,336 points, 45 points behind GPT Image 2 (Medium) at #1, pushing Grok Image 2.0 to #3
On August 8, xAI released Aurora Image 2.0 with precision editing and what xAI describes as improved factuality for real work, and its low-quality setting debuted at #2 on the Text-to-Image Arena with 1320 points, jumping 12 places.
The previous Grok image model dropped to #14 with 1228 points. The model also placed #2 in the Image Edit Arena with 1439 points and is available only through xAI's app, not via API.
Grok Imagine Image 2.0 (Low) debuted at #2 on Text-to-Image Arena with 1,320 points, jumping 12 places; also #2 in Image Edit Arena with 1,439 points
Alibaba reports Qwen-Image-3.0-Pro ranks #1 among Chinese models and #2 among mainstream models on Text-to-Image Arena
Qwen-Image-3.0-Pro enters Text-to-Image Arena at #5 with 1,263 points, up from Qwen-Image-2.0-Pro's #15 at 1,191 points
I send a short email each morning with the stories that moved. If you would rather just read here, that works too.
Subscribe free