MiniMax holds a narrow lead in trader sentiment for the best image-to-video model by end of October, reflecting its recent H3 and post-trained H3 Max variants topping several Artificial Analysis I2V arenas through strong subject consistency, motion realism, and reference handling. ByteDance’s Seedance 2.0 follows closely on multi-image fidelity and native audio, while Google’s Veo 3.1 and Gemini Omni Flash compete on cinematic quality and synchronized sound. Alibaba’s HappyHorse and Wan lines, plus Kling and Luma, keep the field fragmented as no single lab dominates benchmarks. With resolution weeks away, fresh model drops, updated leaderboards, or capability demonstrations could quickly shift the close probabilities among these contenders.
Polymarketデータを参照したAI生成の実験的な要約。これは取引アドバイスではなく、このマーケットの解決方法には一切関係ありません。 · 更新日ByteDance 46%
MiniMax 29%
アリババ/Alibaba-ATH 25%
Google 25%

ByteDance
25%

MiniMax
16%

アリババ/Alibaba-ATH
25%

25%

Lightricks
12%

Black Forest Labs
15%

SpaceXAI
21%

Pixverse
19%

Runway
1%

生成樹
21%

KlingAI
1%

Luma AI
26%

テンセント
23%

Pika
26%

Pruna
30%
ByteDance 46%
MiniMax 29%
アリババ/Alibaba-ATH 25%
Google 25%

ByteDance
25%

MiniMax
16%

アリババ/Alibaba-ATH
25%

25%

Lightricks
12%

Black Forest Labs
15%

SpaceXAI
21%

Pixverse
19%

Runway
1%

生成樹
21%

KlingAI
1%

Luma AI
26%

テンセント
23%

Pika
26%

Pruna
30%
Results from the "Rank" column under the "Image-to-Video Arena" Leaderboard tab at https://arena.ai/leaderboard/image-to-video filtered for "Models" will be used to resolve this market.
Note: Models marked “AutoEval” at the applicable check time will not be considered, regardless of whether they display a rank or score. Models marked “Preliminary” at the applicable check time will be considered.
Models will be ordered primarily by their leaderboard rank at the market’s check time. If two or more models are tied on rank, they will be ordered by which model is listed higher on the leaderboard. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the arena.ai Image-to-Video Arena Leaderboard. If this resolution source is unavailable at check time, this market will remain open until the leaderboard comes back online and will resolve based on the first check after it becomes available. If it becomes permanently unavailable, this market will resolve to "Other".
マーケット開始日: Sep 9, 2026, 3:52 PM ET
リゾルバー
0x69c47De9D...Results from the "Rank" column under the "Image-to-Video Arena" Leaderboard tab at https://arena.ai/leaderboard/image-to-video filtered for "Models" will be used to resolve this market.
Note: Models marked “AutoEval” at the applicable check time will not be considered, regardless of whether they display a rank or score. Models marked “Preliminary” at the applicable check time will be considered.
Models will be ordered primarily by their leaderboard rank at the market’s check time. If two or more models are tied on rank, they will be ordered by which model is listed higher on the leaderboard. If a tie still remains, alphabetical order of company names as listed in this market group will be used as a final tiebreaker (e.g., “Google” would be ranked ahead of “SpaceXAI”). This market will resolve based on the company that occupies first place under this ranking.
The resolution source for this market is the arena.ai Image-to-Video Arena Leaderboard. If this resolution source is unavailable at check time, this market will remain open until the leaderboard comes back online and will resolve based on the first check after it becomes available. If it becomes permanently unavailable, this market will resolve to "Other".
リゾルバー
0x69c47De9D...MiniMax holds a narrow lead in trader sentiment for the best image-to-video model by end of October, reflecting its recent H3 and post-trained H3 Max variants topping several Artificial Analysis I2V arenas through strong subject consistency, motion realism, and reference handling. ByteDance’s Seedance 2.0 follows closely on multi-image fidelity and native audio, while Google’s Veo 3.1 and Gemini Omni Flash compete on cinematic quality and synchronized sound. Alibaba’s HappyHorse and Wan lines, plus Kling and Luma, keep the field fragmented as no single lab dominates benchmarks. With resolution weeks away, fresh model drops, updated leaderboards, or capability demonstrations could quickly shift the close probabilities among these contenders.
Polymarketデータを参照したAI生成の実験的な要約。これは取引アドバイスではなく、このマーケットの解決方法には一切関係ありません。 · 更新日

外部リンクに注意してください。
外部リンクに注意してください。
よくある質問