Compare AI models

Specifications
Parameters
671B72B
Context window
128k128k
API pricing
Cheapest input
$0.256DeepInfra—
Cheapest output
$0.712DeepInfra—
$103StreamLake—
Benchmarks
SWE-Bench Verified
42%60.4%
Overview
CompanyDeepSeekMoonshot AI
Release dateDec 26 2024Jun 17 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Kimi-Dev leads DeepSeek V3 on 1 of the 1 benchmark they both report (SWE-Bench Verified). DeepSeek V3 shipped 173 days before Kimi-Dev, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Kimi-Dev has 72B. Context windows are 128k (DeepSeek V3) vs 128k (Kimi-Dev).

On SWE-Bench Verified, Kimi-Dev leads at 60.4% vs DeepSeek V3 at 42%.