Gemma 4 26B — benchmark results
Google Gemma 4 26B model row. Provider: Google. Released 2026-04-02. Access: API.
Unified ELO 1641 ± 50, rank #390 of 1776 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AgentFloor | 59.9 | TCR (self-reported) | 96.9 |
| SEA-HELM | 70.82 | Mean Score (%) | 93 |
| Wordle Arena | 83.33 | Win Rate (%) | 85.4 |
| Fibble5 Arena | 9.09 | Win Rate (%) | 81.1 |
| TeleResilienceBench | 27.2 | Macro Average CFR (self-reported) | 71.4 |
| DiagnosticIQ | 63.2 | Macro % D.IQ (self-reported) | 65.6 |
| Fibble Arena | 8.33 | Win Rate (%) | 64.1 |
| SLR-Bench - Medium | 23 | Accuracy (%) | 57.3 |
| Polaris-Bench | 47.2 | Overall C (self-reported) | 53.8 |
| Wolfram LLM Benchmarking Project | 41.8 | Correct Functionality (%) | 52.7 |
| SLR-Bench - Hard | 4 | Accuracy (%) | 51.8 |
| AgentCIBench | 67.5 | Leakage (self-reported) | 50 |
Interactive version: theaggregate.ai/model?slug=gemma-4-26b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.