List your company

Text Math Reasoning on GSM8k-ZH

95.9ScoreOpus-4.6
89.34891.04992.7594.451Jun 23, 2026

Evaluation results

MethodLinks
Opus-4.6Parameter Count=≈800B2026.0695.9
Qwen3.5-PlusParameter Count=397B2026.0695.4
GPT-5.4Parameter Count=≈1.5T2026.0694.3
K2.5Parameter Count=≈1T2026.0693.9
Yuvion VL-32BDecoding Mode=Reasoning, Model Scale=32B2026.0693.3
Qwen3-VL-32BDecoding Mode=Instruct, Model Scale=32B2026.0693.2
Qwen3-VL-32BDecoding Mode=Thinking, Model Scale=32B2026.0693.2
Qwen3-VL-8BDecoding Mode=Thinking, Model Scale=8B2026.0691.5
Yuvion VL-32BDecoding Mode=Instruct, Model Scale=32B2026.0691.4
Yuvion VL-8BDecoding Mode=Instruct, Model Scale=8B2026.0689.8
Qwen3-VL-8BDecoding Mode=Instruct, Model Scale=8B2026.0689.7
Yuvion VL-8BDecoding Mode=Reasoning, Model Scale=8B2026.0689.6