List your company

MM Document Understanding on AI2D (test)

91.5ScoreQwen3.5-Plus
81.20483.87786.5589.223Jun 23, 2026

Evaluation results

MethodLinks
Qwen3.5-PlusParameter Count=397B2026.0691.5
K2.5Parameter Count=≈1T2026.0690.5
GPT-5.4Parameter Count=≈1.5T2026.0689.4
Qwen3-VL-32BDecoding Mode=Thinking, Model Scale=32B2026.0688.2
Qwen3-VL-32BDecoding Mode=Instruct, Model Scale=32B2026.0688.1
Opus-4.6Parameter Count=≈800B2026.0687.2
Yuvion VL-32BDecoding Mode=Reasoning, Model Scale=32B2026.0685.7
Qwen3-VL-8BDecoding Mode=Thinking, Model Scale=8B2026.0684.8
Yuvion VL-8BDecoding Mode=Reasoning, Model Scale=8B2026.0684.8
Yuvion VL-32BDecoding Mode=Instruct, Model Scale=32B2026.0684.1
Qwen3-VL-8BDecoding Mode=Instruct, Model Scale=8B2026.0683.9
Yuvion VL-8BDecoding Mode=Instruct, Model Scale=8B2026.0681.6