BlogLLM Benchmark Scores as Published, Side by Side
MotionBench: Results as Published
Every result the catalog holds for MotionBench: 4 models from 4 developers, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Multimodal UnderstandingMaintainer Wenyi Hong and co-authors (Tsinghua University and Zhipu AI)Introduced 2025Data version 2026103001, as of 2026-10-30
MotionBench tests fine-grained motion understanding in video. It has 8,052 manually written multiple-choice questions on clips from the web, public datasets, and synthetic scenes. The six question groups are motion recognition, location-related motion, camera motion, motion-related objects, action order, and repetition count. The reported result is accuracy.
MotionBench
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.2OpenAI, released 2025-12-11 | 64.8% | Moonshot AI, Kimi K2.5 model cardchecked 2026-10-15 | Effort: xhigh |
| Claude Opus 4.5Anthropic, released 2025-11-24 | 60.3% | Moonshot AI, Kimi K2.5 model cardchecked 2026-10-15 | - |
| Gemini 3 ProGoogle, released 2025-11-18 | 70.3% | Moonshot AI, Kimi K2.5 model cardchecked 2026-10-15 | Effort: high |
| Kimi K2.5Moonshot AI, released 2026-01-27 | 70.4% | Moonshot AI, Kimi K2.5 model cardchecked 2026-10-15 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.