All tasks / Models
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.
- MiniMax
- Open weights
- Released 30 Jul 2026
Where it ranks
- SVideo generationModels · Local · 1,159 on Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio)
- AVideo editingModels · Local · 1,125 on Artificial Analysis Video Arena: Video Editing (AA-Video-Editing v1.1, with audio)
Facts
- Vendor
- MiniMax
- Family
- Hailuo (MiniMax H series)
- Released
- 30 Jul 2026
- Weights
- Open: download and run it yourself
- License
- MiniMax H3 Community License Agreement (README links an application form for USA/EU/UK/South Korea)
- Parameters
- 33B
- 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))
- Access
- Free (open weights)
- Last verified
- 18 Oct 2026
Price from artificialanalysis.ai, 18 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Artificial Analysis Video Arena: Image to Video (AA-Video-I2V v1.1, with audio) | 1,140 | MiniMax H3 (768p)AA-Video-I2V v1.1 (with audio) board | 18 Oct 2026 | artificialanalysis.ai |
| Artificial Analysis Video Arena: Image to Video, no audio (AA-Video-I2V-Silent v1.0) | 1,356 | MiniMax H3AA-Video-I2V-Silent v1.0 (no audio) board | 18 Oct 2026 | artificialanalysis.ai |
| Artificial Analysis Video Arena: Text to Video (AA-Video-T2V v2.1, with audio) | 1,159 | MiniMax H3 (768p)AA-Video-T2V v2.1 (with audio) board | 18 Oct 2026 | artificialanalysis.ai |
| Artificial Analysis Video Arena: Text to Video, no audio (AA-Video-T2V-Silent v2.1) | 1,121 | MiniMax H3 (768p)AA-Video-T2V-Silent v2.1 (no audio) board | 18 Oct 2026 | artificialanalysis.ai |
| Artificial Analysis Video Arena: Video Editing (AA-Video-Editing v1.1, with audio) | 1,125 | MiniMax H3 (768p)AA-Video-Editing v1.1 (with audio) board | 18 Oct 2026 | artificialanalysis.ai |
| Artificial Analysis Video Arena: Video Editing, no audio (AA-Video-Editing-Silent v1.1) | 1,129 | MiniMax H3 (768p)AA-Video-Editing-Silent v1.1 (no audio) board | 18 Oct 2026 | artificialanalysis.ai |
| LMArena (Arena) Image-to-Video leaderboard | 1,495 | minimax-h3LMArena official leaderboard dataset (HF lmarena-ai/leaderboard-dataset, config image_to_video, split latest, publish date 2026-09-22); arena.ai page itself returned HTTP 403 | 22 Sep 2026 | datasets-server.huggingface.co |
| LMArena (Arena) Text-to-Video leaderboard | 1,460 | minimax-h3LMArena official leaderboard dataset (HF lmarena-ai/leaderboard-dataset, config text_to_video, split latest, publish date 2026-09-22); arena.ai page itself returned HTTP 403 | 22 Sep 2026 | datasets-server.huggingface.co |
| LMArena (Arena) Video Edit leaderboard | 1,392 | minimax-h3LMArena official leaderboard dataset (HF lmarena-ai/leaderboard-dataset, config video_edit, split latest, publish date 2026-08-27) | 27 Aug 2026 | datasets-server.huggingface.co |
Where you can use it
- SComfyUIWorkflow platforms · Local
- SfalRuntimes & APIs · Cloud
- SFLORAWorkflow platforms · Cloud
- SHailuo AIApps · Cloud
- Sinvideo AIApps · Cloud
- SLumaApps · Cloud
- SMagnific (formerly Freepik)Apps · Cloud
- SPikaApps · Cloud
- SRunwayApps · Cloud
- SSwarmUIApps · Local
- –TopviewTopview AI
- SWanGP (Wan2GP)Apps · Local
Earlier versions
- Hailuo 2.3 · 28 Oct 2025
- Hailuo 02 · 18 Jun 2025
Compare
Sources
- https://huggingface.co/MiniMaxAI/MiniMax-H3
- https://hailuoai.video/
- https://artificialanalysis.ai/video/leaderboard/text-to-video
- https://artificialanalysis.ai/video/leaderboard/image-to-video
- https://datasets-server.huggingface.co/rows?dataset=lmarena-ai/leaderboard-dataset&config=text_to_video&split=latest&offset=0&length=100
- https://datasets-server.huggingface.co/rows?dataset=lmarena-ai/leaderboard-dataset&config=image_to_video&split=latest&offset=0&length=100
- https://github.com/deepbeepmeep/Wan2GP
- https://artificialanalysis.ai/video/leaderboard/video-editing
- https://datasets-server.huggingface.co/rows?dataset=lmarena-ai/leaderboard-dataset&config=video_edit&split=latest&offset=0&length=100