modelgrep

DeepSeek: DeepSeek V4 Pro vs Qwen: Qwen3.7 Max

deepseek logoDeepSeek: DeepSeek V4 Prodeepseek/deepseek-v4-proVSqwen logoQwen: Qwen3.7 Maxqwen/qwen3.7-max

DeepSeek: DeepSeek V4 Pro wins on more metrics (5 of 9), but the right pick depends on what you optimize for — see the breakdown below.

MetricDeepSeek: DeepSeek V4 ProQwen: Qwen3.7 Max
Intelligence Index44.346.0✓
Coding Index59.466.0✓
GPQA Diamond89%92%✓
Design Arena Elo13241325✓
Speed (tokens/sec)56✓50
Latency549ms✓1.3s
Input price /M$0.435✓$1.25
Output price /M$0.870✓$3.75
Context window1.0M✓1M
CapabilitiesReasoningToolsJSONReasoningToolsJSON
DeepSeek: DeepSeek V4 Pro details →Qwen: Qwen3.7 Max details →