Multimodal Reasoning on M3-Bench robot
35.7ME ScoreM3-Agent + StreamMeCo + TMR
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| M3-Agent + StreamMeCo + TMRModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 35.7 | 32.9 | 35.5 | 44 | 27.5 | 34.6 | |
| TimeChat-OnlineModel Category=Streaming Video Models2026.04 | 35.6 | 29.4 | 31.5 | 44.3 | 22.3 | 33.6 | |
| M3-Agent + StreamMeCo + TMRModel Category=Streaming Video Models, Compression Ratio=70%2026.04 | 34.9 | 31.8 | 32.1 | 42.7 | 24.8 | 34.2 | |
| M3-Agent + StreamMeCo + TMRModel Category=Streaming Video Models, Compression Ratio=50%2026.04 | 34.8 | 34.1 | 32.4 | 43.1 | 24.5 | 33.4 | |
| StreamForestModel Category=Streaming Video Models2026.04 | 33.4 | 29.4 | 31.3 | 44.3 | 19 | 32.1 | |
| M3-Agent + StreamMeCoModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 32.9 | 30.6 | 30.9 | 42.5 | 19.6 | 30.7 | |
| M3-Agent + StreamMeCoModel Category=Streaming Video Models, Compression Ratio=50%2026.04 | 32.3 | 28.2 | 29.8 | 41.4 | 21.1 | 30.6 | |
| M3-Agent + TimeChat-MemoryModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 31.8 | 36.5 | 31.7 | 42.2 | 18.7 | 30.2 | |
| VideoChat-OnlineModel Category=Streaming Video Models2026.04 | 31.7 | 24.7 | 30.5 | 43.1 | 17.1 | 29.9 | |
| M3-Agent + MemoryLLMModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 31.4 | 34.1 | 30.9 | 41.4 | 18 | 28.9 | |
| M3-AgentModel Category=Streaming Video Models2026.04 | 30.9 | 29.4 | 29.6 | 41.1 | 22.6 | 30.3 | |
| M3-Agent + ClusteringModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 30 | 32.9 | 30 | 42.5 | 20.8 | 29.6 | |
| M3-Agent + DARTModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 29.5 | 24.7 | 25.8 | 40.1 | 21.7 | 29.1 | |
| M3-Agent + StreamMeCoModel Category=Streaming Video Models, Compression Ratio=70%2026.04 | 29.1 | 30.6 | 26.9 | 40.3 | 19.9 | 28.4 | |
| M3-Agent + RandomModel Category=Streaming Video Models, Compression Ratio=30%2026.04 | 28.9 | 29.4 | 27.7 | 40.5 | 18 | 28.5 | |
| MA-LMMModel Category=Streaming Video Models2026.04 | 25.6 | 23.4 | 22.7 | 39.1 | 14.4 | 24.4 | |
| Flash-VStreamModel Category=Streaming Video Models2026.04 | 21.6 | 19.4 | 19.3 | 24.3 | 14.1 | 19.4 | |
| Gemini-GPT4o-HybridModel Category=Closed-source MLLMs2026.04 | 21.3 | 25.5 | 22.7 | 28.8 | 23.1 | 24 | |
| Gemini-AgentModel Category=Closed-source MLLMs2026.04 | 15.8 | 17.1 | 15.3 | 20 | 15.5 | 16.9 | |
| MovieChatModel Category=Streaming Video Models2026.04 | 13.3 | 9.8 | 12.2 | 15.7 | 7 | 11.2 | |
| GPT-4oModel Category=Closed-source MLLMs2026.04 | 9.3 | 9 | 8.4 | 10.2 | 7.3 | 8.5 | |
| Gemini-1.5-ProModel Category=Closed-source MLLMs2026.04 | 6.5 | 7.5 | 8 | 9.7 | 7.6 | 8 | |
| Qwen2.5-VL-7b-InstructModel Category=Open-source MLLMs2026.04 | 2.9 | 3.8 | 3.6 | 4.6 | 3.4 | 3.4 | |
| Qwen2.5-Omni-7bModel Category=Open-source MLLMs2026.04 | 2.1 | 1.4 | 1.5 | 1.5 | 2.1 | 2 |