Compare AI models

Specifications
Parameters
671B1T
Context window
128k256k
API pricing
Cheapest input
$0.256DeepInfra$0.49AtlasCloud
Cheapest output
$0.712DeepInfra$2.50AtlasCloud
$103StreamLake$223AtlasCloud · int4
Benchmarks
SWE-Bench Verified
42%76.8%
LiveCodeBench
40.5%85%
GPQA Diamond
59.1%87.6%
Overview
CompanyDeepSeekMoonshot AI
Release dateDec 26 2024Jan 27 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Kimi K2.5 leads DeepSeek V3 on 3 of the 3 benchmarks they both report (SWE-Bench Verified, LiveCodeBench, GPQA Diamond). DeepSeek V3 shipped 397 days before Kimi K2.5, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Kimi K2.5 has 1T. Context windows are 128k (DeepSeek V3) vs 256k (Kimi K2.5).

On SWE-Bench Verified, Kimi K2.5 leads at 76.8% vs DeepSeek V3 at 42%. On LiveCodeBench, Kimi K2.5 leads at 85% vs DeepSeek V3 at 40.5%. On GPQA Diamond, Kimi K2.5 leads at 87.6% vs DeepSeek V3 at 59.1%.