Chinese AI video models compared: MiniMax H3 takes first place overall
In the latest SuperCLUE I2V (image-to-video) model overall ranking, MiniMax (Xiyu Technology) H3 took first place in China with an overall score of 71.46, and its application capability score of 75.82 was the highest on the entire list. Alibaba's Wan 3.0 ranked second with 66.63, and its basic capability score of 68.51 was slightly higher than that of MiniMax H3; ByteDance's Seedance 2.5 and Kuaishou's Kling V3 Pro tied for third with 64.63 and 64.06. The ranking was updated on October 23; overseas models are included only for reference and are not counted in the official ranking.
Body · AI translation
Chinese AI Video Models Compared: MiniMax H3 Takes First Place Overall
Kuaikeji Download the App | Exclusive First Look October 25, 2026, 17:17:34 | From Henan
Kuaikeji News, October 25 — The latest SuperCLUE I2V image-to-video model overall leaderboard has been released, and MiniMax Technology's MiniMax H3 has taken first place with a total score of 71.46, running ahead of all domestic AI video models.
This leaderboard was updated on October 23. There is a key point in the rules: overseas models are used only for comparison and are not included in the official rankings, and models whose score gap is within 1 point are counted as tied.
MiniMax H3 not only has a high total score; breaking it down into two capabilities, its basic capability scored 67.10, while its application capability surged to 75.82, the highest application capability score in the table.
Ranking second domestically is Alibaba's Wan 3.0, with a total score of 66.63. Its basic capability performance is strong, scoring 68.51, slightly higher than MiniMax H3, but its application capability held it back somewhat, at only 64.75.
Third place is a tie between two models: ByteDance's Seedance 2.5 scored 64.63, and Kuaishou's Kling V3 Pro scored 64.06.
Further down, AiShi Technology's PixVerse V6 and Shengshu Technology's Vidu Q3 Pro tied for 4th place.
Several overseas models are also included in the leaderboard for reference: Google's Gemini Omni 1.1 Flash has a total score of 67.38, and X.AI's Grok Imagine Video 1.5 scored 65.79. Neither of these is counted in the official rankings.
The notes at the end also make it clear that several models are subject to rule restrictions. For example, Seedance 2.5, Gemini Omni 1.1 Flash, and PixVerse V6 encounter constraints such as restrictions on real people appearing on camera, generation duration limits, and content moderation; some questions could not produce results normally and were directly counted as 0 points, which affects their final scores.
However, this is only a benchmark score. A high score does not mean the model works well in all scenarios. When actually creating content, everyone should still choose a suitable model according to their own needs.
"Special Statement: The above works (including videos, pictures, and audio if any) are uploaded and published by users of Dafeng Hao, a self-media platform under Phoenix Net. This platform only provides information storage space services.
Notice: The content above (including the videos, pictures and audios if any) is uploaded and posted by the user of Dafeng Hao, which is a social media platform and merely provides information storage space services."