← All models
Compare

Gemini 3.1 Pro vs Kimi K2.5

Add, remove, or swap models to compare them side by side.

Gemini 3.1 ProKimi K2.5
Gemini 3.1 ProGoogle DeepMind · 2 in the newsKimi K2.5Moonshot · 10 in the news
Scores
Intelligence (ECI)156148
Coding7146
Math7653
Reasoning & Knowledge9637
Agentic & Tools8839
Specifications
DeveloperGoogle DeepMindMoonshot
FamilyGeminiKimi
ReleasedFeb 19, 2026Feb 2, 2026
Parameters1T
AvailabilityAPI accessOpen weights (unrestricted)
Context window262K
Price — $/M input$0.38
Price — $/M output$2.02
Inputstext, image
Outputstext
Benchmarks
AIME 2024/202596%92%
APEX34%14%
ARC-AGI98%65%
ARC-AGI-277%12%
FrontierMath37%28%
FrontierMath Tier 417%4%
GPQA Diamond94%88%
GSO (code optimization)23%
Humanity's Last Exam46%24%
METR task horizon6.4 h
SimpleBench80%47%
SimpleQA Verified77%34%
SWE-bench Verified76%74%
Terminal-Bench80%43%
WebDev Arena14611431
WeirdML72%46%

Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.

Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.