List your company

MM STEM & Puzzle on MathVista mini

89.1ScoreK2.5
76.20479.55282.986.248Jun 23, 2026

Evaluation results

MethodLinks
K2.5Parameter Count=≈1T2026.0689.1
Qwen3.5-PlusParameter Count=397B2026.0686.7
Qwen3-VL-32BDecoding Mode=Thinking, Model Scale=32B2026.0683.9
Qwen3-VL-32BDecoding Mode=Instruct, Model Scale=32B2026.0682.4
Yuvion VL-32BDecoding Mode=Reasoning, Model Scale=32B2026.0681.4
Yuvion VL-32BDecoding Mode=Instruct, Model Scale=32B2026.0680.5
Qwen3-VL-8BDecoding Mode=Thinking, Model Scale=8B2026.0679.6
Yuvion VL-8BDecoding Mode=Reasoning, Model Scale=8B2026.0679.2
Opus-4.6Parameter Count=≈800B2026.0677.6
GPT-5.4Parameter Count=≈1.5T2026.0677.2
Yuvion VL-8BDecoding Mode=Instruct, Model Scale=8B2026.0677
Qwen3-VL-8BDecoding Mode=Instruct, Model Scale=8B2026.0676.7