Video Motion & Perception on TOMATO
95.2AccuracyHuman
Evaluation results
| Method | Links | |
|---|---|---|
| HumanThinking with Tracking=false2026.06 | 95.2 | |
| HumanThinking with Tracking=true2026.06 | 95.2 | |
| Seed2.0 ProThinking with Tracking=true2026.06 | 65.3 | |
| Gemini-3-FlashThinking with Tracking=true2026.06 | 64 | |
| Seed1.8Thinking with Tracking=true2026.06 | 61 | |
| Gemini-3-FlashThinking with Tracking=false2026.06 | 60.8 | |
| Seed1.8Thinking with Tracking=false2026.06 | 60.8 | |
| Seed2.0 ProThinking with Tracking=false2026.06 | 59.9 | |
| Gemini-3-ProThinking with Tracking=false2026.06 | 59.6 | |
| Seed2.0 LiteThinking with Tracking=true2026.06 | 59.2 | |
| Seed2.0 LiteThinking with Tracking=false2026.06 | 57.3 | |
| Seed2.0 MiniThinking with Tracking=true2026.06 | 51.3 | |
| Seed2.0 MiniThinking with Tracking=false2026.06 | 47.4 |