Compare AI models

Specifications
Parameters
671B1T
Context window
128k256k
API pricing
Cheapest input
$0.256DeepInfra$0.60Novita
Cheapest output
$0.712DeepInfra$2.50Novita
$103StreamLake$245Novita · bf16
Benchmarks
SWE-Bench Verified
42%71.3%
Overview
CompanyDeepSeekMoonshot AI
Release dateDec 26 2024Nov 6 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Kimi K2 Thinking leads DeepSeek V3 on 1 of the 1 benchmark they both report (SWE-Bench Verified). DeepSeek V3 shipped 315 days before Kimi K2 Thinking, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Kimi K2 Thinking has 1T. Context windows are 128k (DeepSeek V3) vs 256k (Kimi K2 Thinking).

On SWE-Bench Verified, Kimi K2 Thinking leads at 71.3% vs DeepSeek V3 at 42%.