MiniMax-H3 Tops Video Editing Leaderboard, Ties for 2nd in Text-to-Video
bdsqlsz · x · 2026-07-31
MiniMax officially confirmed that its latest video model, MiniMax-H3, is open-source. According to the Artificial Analysis leaderboard, the model shows top-tier performance:
- Video Editing (With Audio): Ranked #1 globally.
- Text to Video (With Audio): Tied for #2 globally alongside Google Gemini Omni Flash.
It also secured the #2 spot in Image to Video (No Audio), demonstrating comprehensive capabilities.
Related event: MiniMax-H3 Open-Sourced, Topping AI Video Editing Leaderboard(2 posts)→
More from Multimodal
- Creating Fantasy Short Films with Hailuo and Midjourney — beholdersai · 2026-07-31
- MiniMax H3 Video Model Launches on Topview at 30% of Seedance 2.0's Price — iamfakhrealam · 2026-07-31
- AI Video Generation Tackles Soccer Simulation with Red Cards and Injuries — talkaboutdesign · 2026-07-31
- 4B Model Arko-T Beats GPT-5 in Text-to-CAD Generation — 机器之心 · 2026-07-31
- Elon Musk Announces xAI's New Image Generation Feature Grok Imagine — elonmusk · 2026-07-31
- Beacon: Agentic Visual Reasoning Model Improves Tool Adaptiveness and Effectiveness — KlingTeam · 2026-07-31