Compare
Gemma 4 26B A4B vs Kimi K2.5
Gemma 4 26B A4B (Google DeepMind) and Kimi K2.5 (Moonshot) compared on benchmarks, pricing, context window, and use-case rankings.
- Kimi K2.5 ranks higher for coding (44th vs 26th percentile).
Scores
Intelligence (ECI)—148
Coding2644
Math—53
Reasoning & Knowledge436
Agentic & Tools—39
Specifications
DeveloperGoogle DeepMindMoonshot
FamilyGemmaKimi
ReleasedApr 2, 2026Feb 2, 2026
Parameters—1T
AvailabilityOpen weights (unrestricted)Open weights (unrestricted)
Context window—262K
Price — $/M input—$0.38
Price — $/M output—$2.02
Inputs—text, image
Outputs—text
Benchmarks
WebDev Arena13591431
WeirdML35%46%
AIME 2024/2025—92%
APEX—14%
ARC-AGI—65%
ARC-AGI-2—12%
FrontierMath—28%
FrontierMath Tier 4—4%
GPQA Diamond—88%
Humanity's Last Exam—24%
SimpleBench—47%
SimpleQA Verified—34%
SWE-bench Verified—74%
Terminal-Bench—43%
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.
Frequently asked questions
Which is better for coding, Gemma 4 26B A4B or Kimi K2.5?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Kimi K2.5 ranks higher — 44th vs 26th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.