List your company

Video Reasoning on TwiFF-Bench

3.11Avg CoTFUTURE-L1-RL
1.99722.28612.5752.8639Feb 11, 2026Mar 1, 2026Mar 20, 2026Apr 8, 2026Apr 27, 2026May 16, 2026Jun 4, 2026

Evaluation results

MethodLinks
FUTURE-L1-RLSize=8B2026.063.11——————2.973.04
TwiFFSize=7B2026.022.952.812.433.243.043.323.142.62—
TwiFF-2.7MSize=7B2026.062.95——————2.622.79
TwiFF-LiteSize=7B2026.022.92.762.343.133.033.343.112.55—
TwiFF-300KSize=7B2026.062.9——————2.552.73
Qwen3VLSize=8B2026.022.842.782.312.882.83.12.762.44—
Zero-ShotSize=8B2026.062.75——————2.142.44
FUTURE-L1-SFTSize=8B2026.062.62——————2.422.52
DeepEyesSize=7B2026.022.542.492.092.762.692.572.172.2—
DeepEyesSize=7B2026.062.54——————2.22.37
Qwen2.5VLSize=7B2026.022.462.41.642.571.772.641.411.63—
Qwen2.5-VLSize=7B2026.062.46——————1.632.05
InternVL3.5Size=8B2026.022.352.241.762.52.272.771.831.85—
InternVL3.5Size=8B2026.062.35——————1.852.1
BagelSize=7B2026.022.292.21.722.452.12.612.221.85—
BagelSize=7B2026.062.29——————1.852.07
Zebra-COTSize=7B2026.022.272.161.222.612.12.471.631.41—
ThinkMorphSize=7B2026.022.212.151.32.482.382.211.43—
Janus-ProSize=7B2026.022.041.9512.141.362.380.931.04—
Janus-ProSize=7B2026.062.04——————1.041.54