All tasks / Comparisons
MiniMax H3 vs Seedance 2.5
Compared only on public benchmark results, across the 2 tasks where both are listed: they share a tier in 2.
MiniMax H3
MiniMax's omni-modal video model with native stereo audio, up to 2K and 15 s; weights released, successor to Hailuo 2.3.
- Vendor
- MiniMax
- Weights
- Open: run it yourself
- Released
- 30 Jul 2026
- API price
- $0.08 per second of video (768p with audio, $4.80/min as listed on the AA text-to-video board ($5.40/min for image-to-video))
- License
- MiniMax H3 Community License Agreement (README links an application form for USA/EU/UK/South Korea)
- Last verified
- 18 Oct 2026
Seedance 2.5
ByteDance Seed's video model line with native audio, offered in Dreamina/CapCut and via API.
- Vendor
- ByteDance (Seed)
- Weights
- Closed: API and apps only
- Released
- 31 Jul 2026
- API price
- $0.57 per second of video ($34.12/min as listed on the AA text-to-video board (with audio; resolution not stated); $24.60/min for image-to-video)
- License
- proprietary
- Last verified
- 18 Oct 2026
Task by task
| Task | MiniMax H3 | Seedance 2.5 | Higher tier |
|---|---|---|---|
| Video generation Models | S1,159 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio) · Local | S1,160 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio) · Cloud | Same tier |
| Video editing Models | A1,125 on Artificial Analysis Video Arena: Video Editing (AA-Video-Editing v1.1, with audio) · Local | A1,155 on Artificial Analysis Video Arena: Video Editing (AA-Video-Editing v1.1, with audio) · Cloud | Same tier |
A tier spans a range of scores, and small gaps on a leaderboard are often within its margin of error, so only a different tier counts as ahead. How we rank.
Same benchmark, both results
| Benchmark | MiniMax H3 | Seedance 2.5 | Snapshot |
|---|---|---|---|
| Artificial Analysis Video Arena: Image to Video (AA-Video-I2V v1.1, with audio) Elo (pairwise human preference), higher is better | 1,140 | 1,138 | 18 Oct 2026 |
| Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio) Elo (pairwise human preference), higher is better | 1,159 | 1,160 | 18 Oct 2026 |
| Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1) Elo (pairwise human preference), higher is better | 1,121 | 1,105 | 18 Oct 2026 |
| Artificial Analysis Video Arena: Video Editing (AA-Video-Editing v1.1, with audio) Elo (pairwise human preference), higher is better | 1,125 | 1,155 | 18 Oct 2026 |
| Artificial Analysis Video Arena: Video Editing, no audio (AA-Video-Editing-Silent v1.1) Elo (pairwise human preference), higher is better | 1,129 | 1,146 | 18 Oct 2026 |
| LMArena (Arena) Image-to-Video leaderboard Arena score (Bradley-Terry Elo-style rating from pairwise votes), higher is better | 1,495 | 1,477 | 22 Sep 2026 |
| LMArena (Arena) Text-to-Video leaderboard Arena score (Bradley-Terry Elo-style rating from pairwise votes), higher is better | 1,460 | 1,474 | 22 Sep 2026 |
| LMArena (Arena) Video Edit leaderboard Arena score (Bradley-Terry Elo-style rating from pairwise votes), higher is better | 1,392 | 1,410 | 27 Aug 2026 |