DeepSeek V4 Pro vs Kimi K2 Thinking

DeepSeek V4 Pro edges ahead on overall intelligence (RunFree 72.4 vs 45.2). Here's how they stack up on benchmarks, price and specs.

DeepSeek V4 Pro
DeepSeek
Kimi K2 Thinking
MoonshotAI

Shared benchmarks (6)

Loading chart…
DeepSeek V4 ProKimi K2 Thinking
RunFree Score72.445.2
Blended price / 1M$0.54$1.07
Input / 1M$0.43$0.60
Output / 1M$0.87$2.50
Context window1.0M262K
Max output384K262K
Reasoning modelYesYes
GPQA Diamond90.1%84.5%
Humanity's Last Exam37.7%23.9%
LiveCodeBench93.5%83.1%
MMLU-Pro87.5%84.6%
SWE-Bench Verified80.6%71.3%
Terminal-Bench67.9%47.1%
Wins120

A blank means the metric is tied or one model has no verified data. Prices are blended 3:1 input:output.

Compare more: DeepSeek V4 Pro card · Kimi K2 Thinking card · full leaderboard