List your company

Reasoning on MME-r reasoning

722.5ScoreQwen3-VL-8B-Thinking
-1.236186.657374.55562.443Feb 18, 2026Mar 7, 2026Mar 24, 2026Apr 10, 2026Apr 27, 2026May 14, 2026Jun 1, 2026

Evaluation results

MethodLinks
Qwen3-VL-8B-ThinkingBackbone=Qwen3-VL-8B, Strategy=LongCoT (Thinking)2026.02722.5
SAPBackbone=Qwen3-VL-8B, Strategy=SAP2026.02689.9
Qwen3-VL-8B-InstructBackbone=Qwen3-VL-8B, Strategy=Instruct2026.02656.7
MiMo-VL-7B-SFTRun=+TRON2026.0646.46
Qwen3-VL-4B-InstructRun=+TRON2026.0645.88
MiMo-VL-7B-SFTRun=Base2026.0645.29
Qwen3-VL-4B-InstructRun=Base2026.0642.85
Qwen2.5-VL-7B-InstructRun=+TRON2026.0628.62
Qwen2.5-VL-7B-InstructRun=Base2026.0626.6