Compare
Gemma 4 26B A4B vs Kimi K3
Gemma 4 26B A4B (Google DeepMind) and Kimi K3 (Moonshot) compared on benchmarks, pricing, context window, and use-case rankings.
- Kimi K3 ranks higher for coding (99th vs 27th percentile).
| Attribute | Gemma 4 26B A4B | Kimi K3 |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | — | 157 |
| Coding | 27 | 99 |
| Math | — | 81 |
| Reasoning & Knowledge | 19 | 78 |
| Agentic & Tools | 14 | 84 |
| Specifications | ||
| Developer | Google DeepMind | Moonshot |
| Family | Gemma | Kimi |
| Released | Apr 2, 2026 | Jul 16, 2026 |
| Parameters | — | 2.8T |
| Availability | Open weights (unrestricted) | Open weights (non-commercial) |
| Context window | — | 1M |
| Price — $/M input | — | $2.80 |
| Price — $/M output | — | $14.00 |
| Inputs | — | text, image, video |
| Outputs | — | text |
| Benchmarks | ||
| GPQA Diamond | 79% | 93% |
| Humanity's Last Exam | 19% | 47% |
| SciCode | 40% | 59% |
| WebDev Arena | 1359 | 1682 |
| WeirdML | 35% | 83% |
| τ²-bench | 44% | — |
| AIME 2024/2025 | — | 97% |
| APEX | — | 39% |
| ARC-AGI | — | 95% |
| ARC-AGI-2 | — | 60% |
| SimpleQA Verified | — | 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 K3?
Across coding benchmarks like SWE-bench Verified and Terminal-Bench, Kimi K3 ranks higher — 99th vs 27th 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.