Future Video Reasoning on FutureBench 1-Hop
83.2AccuracyFUTURE-L1
Evaluation Results
| Method | Links | |
|---|---|---|
| FUTURE-L1Size=8B, Method=SFT+RL, Frames=322026.06 | 83.2 | |
| Video-CoESize=7B, Method=SFT+RL, Frames=322026.06 | 80.9 | |
| FUTURE-L1Size=8B, Method=SFT, Frames=322026.06 | 70.5 | |
| Video-o3Size=7B, Method=SFT+RL, Frames=322026.06 | 68.2 | |
| Video-R1Size=7B, Method=SFT+RL, Frames=322026.06 | 67.6 | |
| Text-Only SFTSize=8B, Method=SFT, Frames=322026.06 | 67.6 | |
| NEPSize=7B, Method=SFT+RL, Frames=322026.06 | 66.2 | |
| Qwen3-VLSize=30B-A3B, Method=Zero-Shot, Frames=322026.06 | 65.3 | |
| Qwen3-VL-InstructSize=8B, Method=Zero-Shot, Frames=322026.06 | 64.2 | |
| VideoAuto-R1Size=8B, Method=SFT+RL, Frames=322026.06 | 63.6 | |
| Video-RFTSize=7B, Method=SFT+RL, Frames=322026.06 | 62.4 | |
| GPT-4oSize=–, Method=Zero-Shot, Frames=322026.06 | 61.9 | |
| GPT-5Size=–, Method=Zero-Shot, Frames=322026.06 | 59.6 | |
| MiMo-VLSize=7B, Method=Zero-Shot, Frames=322026.06 | 59 | |
| SwimBirdSize=8B, Method=SFT, Frames=322026.06 | 59 | |
| Qwen2.5-VL-InstructSize=7B, Method=Zero-Shot, Frames=322026.06 | 57.2 | |
| Qwen2.5-VL-InstructSize=72B, Method=Zero-Shot, Frames=322026.06 | 55.5 | |
| InternVL3Size=8B, Method=Zero-Shot, Frames=322026.06 | 54.3 | |
| LLaVA-NeXT-VideoSize=7B, Method=Zero-Shot, Frames=322026.06 | 48.8 | |
| MonetSize=7B, Method=SFT+RL, Frames=322026.06 | 46.8 | |
| GLM-4.1VSize=9B, Method=Zero-Shot, Frames=322026.06 | 29.9 | |
| LVRSize=7B, Method=SFT+RL, Frames=322026.06 | 22.5 |