← All Models

DeepSeek V3.2 vs Kimi K2

DeepSeek V3.2 leads in 3 of 8 categories

Arena ELO

Tie

Cheaper

DeepSeek V3.2

Faster

DeepSeek V3.2

Detailed Comparison

CategoryDeepSeek V3.2Kimi K2
ProviderDeepSeek Moonshot
Context Window128K 128K
Input Price ($/1M)$0.27 WIN$0.55
Output Price ($/1M)$1.1 WIN$2.2
Arena ELO1380 1380
Codingunknown 65.8% SWE-bench
Reasoningunknown 60.2% BrowseComp
Speed (tokens/sec)~80 t/s WIN~50 t/s
Open SourceYes Yes

Frequently Asked Questions

Which is cheaper, DeepSeek V3.2 or Kimi K2?

DeepSeek V3.2 is approximately 50% cheaper than Kimi K2 per million tokens.

Which is better for coding, DeepSeek V3.2 or Kimi K2?

DeepSeek V3.2 scores unknown on coding benchmarks, while Kimi K2 scores 65.8% SWE-bench.

Which has a larger context window?

DeepSeek V3.2 supports 128K context, while Kimi K2 supports 128K.

Is DeepSeek V3.2 or Kimi K2 open source?

DeepSeek V3.2 is open source. Kimi K2 is open source.

Which ranks higher on the Chatbot Arena leaderboard?

Kimi K2 ranks higher with ELO 1380 vs 1380.