Artificial Analysis · Intelligence Index · ReasoningBenchmarks, as Artificial Analysis ran them- 80%GPQA Diamond
- 15%Humanity's Last Exam
- 31%Terminal-Bench Hard
- 79%LiveCodeBench
- 88%AIME 2025
- 34%τ²-Bench
- 85%MMLU-Pro
- 54%IFBench
- 72%Long context reasoning
- Per 1M tokens
- $0.28 in · $0.42 out
- Context
- 164K tokens
LMArena · Text129thof 416· 37th of 226 open models
Rated by versionDeepSeekDeepSeek-V3.2 rates level with it.
DeepSeek V3.2 Exp · 1,422 (1,415–1,428) · 129th
Against list priceRating for the price
DeepSeek V3.2 Exp · 1,422 · $0.31 per 1M tokens ($0.27 in, $0.41 out) · 164K context
DeepSeek V3.2 Exp · 1,422 · $0.31 per 1M tokens ($0.27 in, $0.41 out) · 164K context
LMArena, CC BY 4.0, 15 October 2026 · prices OpenRouter, list price blended 3:1, read 18 October 2026 · Artificial Analysis, read 18 October 2026
DeepSeek V3.2 Exp