Vibeleaderboard
← All Intel
Intel / post

Qwen-Image-3.0-Pro debuts top 10 on Artificial Analysis image arenas

Source
x.com
Date
Artificial Analysis@ArtificialAnlys

Alibaba's Qwen-Image-3.0-Pro lands at #6 on the Artificial Analysis Image Editing Leaderboard and #9 on the Text to Image Leaderboard, up 83 and 48 Elo points on the previous Qwen Pro generation Qwen-Image-3.0 is the third generation of Alibaba's Qwen image family, spanning two models: Qwen-Image-3.0-Pro, the flagship with agent-based prompt rewriting, and Qwen-Image-3.0, which Alibaba positions as the same capabilities with faster generation for high-volume use. Qwen centers the model’s capabilities on realism and information density: prompts of up to 4.5k tokens, text rendered precisely at sizes as small as 10 pixels, and native rendering of 12 languages. In the Artificial Analysis Image Arena, Qwen-Image-3.0-Pro debuts at #6 on the Image Editing Leaderboard, behind only Microsoft's MAI-Image-2.5-Pro and MAI-Image-2.5, Reve 2.1, and OpenAI's GPT Image 2 and GPT Image 1.5. On the Text to Image Leaderboard it debuts at #9, just behind Google's Nano Banana 2 Lite and ahead of ByteDance's Seedream 5.0 Pro. This is a leap over the previous generation: Qwen-Image-3.0-Pro sits 83 Elo points above Qwen-Image-2.0-Pro in Image Editing (#34 to #6) and 48 points above it in Text to Image…

Read the full post on X

Context

Artificial Analysis, which runs blind-vote arenas for AI image models, placed Alibaba's Qwen-Image-3.0-Pro sixth on its image-editing leaderboard and ninth on its text-to-image leaderboard when the model debuted there in August. Its predecessor from the Qwen team, Qwen-Image-2.0-Pro, had ranked 34th for editing and 12th for generation.

The rankings come from head-to-head votes. Evaluators compare two images made from the same prompt without knowing which model made which, and Artificial Analysis turns those choices into Elo scores, the rating system used in chess, in which a model climbs by winning more of its matchups. The reported gains of 83 Elo points in editing and 48 in text-to-image therefore mean evaluators picked the new model's images noticeably more often. They are rating differences, not percentage improvements in image quality.

The launch positions were a snapshot. When checked on October 30, the text-to-image board listed Qwen-Image-3.0-Pro 17th and the image-editing board listed it 21st. On editing it still sat about 33 points above Qwen Image 2.0 Pro, with 95% intervals that do not overlap. The gain over the previous generation held up; the top-ten placing did not.

More from Artificial Analysis
Recommended reads
Comments

Checking sign-in…

Loading comments…