DeepSeek-V3.2-Exp vs Kimi K2-Thinking-0905: Specs & Benchmark Comparison

CharacteristicDeepSeek-V3.2-ExpKimi K2-Thinking-0905
CompanyDeepSeekMoonshot AI
Release DateSeptember 28, 2025September 4, 2025
Parameters685B1.0T
MultimodalNoNo
Context (input)164K262K
Context (output)66K66K
Input Price / 1M$0.27$0.60
Output Price / 1M$0.41$2.40
Average Score0.80.9
Benchmarks
AIME 20250.91.0
GPQA0.80.8
MMLU-Pro0.80.8

Visual Benchmark Comparison

DeepSeek-V3.2-Exp
Kimi K2-Thinking-0905
AIME 20250.9 vs 1.0
0.9
1.0
GPQA0.8 vs 0.8
0.8
0.8
MMLU-Pro0.8 vs 0.8
0.8
0.8

Verdict

DeepSeek-V3.2-Exp leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: DeepSeek-V3.2-Exp — 0.8, Kimi K2-Thinking-0905 — 0.9.

API Cost

DeepSeek-V3.2-Exp is 4.4x cheaper: input $0.27/1M vs $0.60/1M tokens.

Context Window

Kimi K2-Thinking-0905 supports a larger context: 262K vs 164K tokens.

Recency

DeepSeek-V3.2-Exp is newer: released 9/28/2025 vs 9/4/2025.

More About These Models

Related Comparisons

The DeepSeek-V3.2-Exp and Kimi K2-Thinking-0905 comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the DeepSeek-V3.2-Exp or Kimi K2-Thinking-0905 page. See also the complete list of AI model comparisons.