MiniMax-H3 Scores Just 41.97% on a New Omni-Modal Physical World Reasoning Benchmark
Haoyu Zhao · hf · 2026-09-18
A new evaluation framework tests whether the omni-modal generative model MiniMax-H3 can actually reason about the physical world.
- The benchmark is built around four dimensions of physical-world reasoning and exploits omni-modal inputs: implicit prompts with multiple frames, audio-image, prefix-videos, and audio-video combinations—each modality offers only partial evidence, forcing joint cross-modal inference of latent event states and future dynamics.
- Across 517 instances, MiniMax-H3 achieves a 41.97% overall success rate. Video-based Decision Reasoning is strongest at 56.00%, while Audio-based Disambiguation Reasoning reaches only 27.40%.
- The takeaway: effective multimodal integration remains the key bottleneck for unlocking omni-modal inputs. The project is open-sourced at github.com/gulucaptain/MiniMax-H3-Reason.
More from Multimodal
- Free course covers the 4 hard problems of production generative media agents — Al_Grigor · 2026-11-04
- PixelBench: testing multimodal models pixel-by-pixel instead of with QA — bdsqlsz · 2026-11-04
- Kitten-TTS 2 offers free voice cloning on laptops, no GPU needed — solyarisoftware · 2026-11-04
- "We'll Take Your RAM": a fully local-hardware cyberpunk meme from 2030 San Francisco — Wonderful-Yak-3348 · 2026-11-04
- Openfilm: open-source AI video agent turns coding agents into video makers — solyarisoftware · 2026-11-04
- Microsoft reopens VibeVoice: 10-second voice cloning, 90-minute local TTS — solyarisoftware · 2026-11-04