Video Question Answering on TOMATO
35.58ScoreSeed 2.0 Pro
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Seed 2.0 ProAccess=Proprietary2026.07 | 35.58 | — | |
| Gemini 3.0 ProAccess=Proprietary2026.07 | 35.31 | — | |
| Qwen3-VL-235B-A22B-InstructAccess=Open-source2026.07 | 23.05 | — | |
| CuReAccess=Open-source2026.07 | 22.91 | — | |
| Qwen3.6-35B-A3BAccess=Open-source2026.07 | 21.9 | — | |
| OwlCap-7BAccess=Open-source2026.07 | 21.77 | — | |
| Qwen3-VL-30B-A3B-InstructAccess=Open-source2026.07 | 21.5 | — | |
| Tarsier2-7BAccess=Open-source2026.07 | 11.39 | — | |
| InternVL-2.5Params Enc / LLM=300M/7B, Evaluation Protocol=zero-shot2025.06 | — | 29.4 | |
| PLM 8BParams Enc / LLM=1B/8B, Evaluation Protocol=zero-shot2025.06 | — | 33.2 | |
| Qwen2.5VLParams Enc / LLM=1B/7B, Evaluation Protocol=zero-shot2025.06 | — | 24.6 | |
| Qwen2VLParams Enc / LLM=675M/7B, Evaluation Protocol=zero-shot2025.06 | — | 31.5 | |
| V-JEPA 2Params Enc / LLM=1B/8B, Evaluation Protocol=zero-shot2025.06 | — | 40.3 |