modelgrep

DeepSeek: DeepSeek V3.1 vs Qwen: Qwen3 Next 80B A3B Thinking

DDeepSeek: DeepSeek V3.1deepseek/deepseek-chat-v3.1VSQQwen: Qwen3 Next 80B A3B Thinkingqwen/qwen3-next-80b-a3b-thinking

Qwen: Qwen3 Next 80B A3B Thinking wins on more metrics (5 of 9), but the right pick depends on what you optimize for — see the breakdown below.

MetricDeepSeek: DeepSeek V3.1Qwen: Qwen3 Next 80B A3B Thinking
Intelligence Index28.1✓26.7
Coding Index28.4✓19.5
GPQA Diamond74%76%✓
Design Arena Elo1167✓—
Speed (tokens/sec)110179✓
Latency378ms✓1.1s
Input price /M$0.210$0.098✓
Output price /M$0.790$0.780✓
Context window164K262K✓
CapabilitiesReasoningToolsJSONReasoningToolsJSON
DeepSeek: DeepSeek V3.1 details →Qwen: Qwen3 Next 80B A3B Thinking details →