Gemini 4 Neon (Medium): benchmark results
Provider: Google. Released 2026-11-18. Access: API.
Aggregate Rating 1962 ± 12, rank #20 of 2286 rated models, from 112 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA MKA - Output File Type - spreadsheet - Regression WSR | 97.89 | Weighted Satisfaction Rate (%) | 100 |
| AI Coding Daily (Antigravity) - React-TS Code Quality | 18 | React-TS Code Quality (max 20) points, LLM-judged rubric sco | 100 |
| AA GDP.pdf | 32.8 | All Criteria Pass Rate (%) | 99.5 |
| AA GDP.pdf - Criterion Pass Rate | 82.76 | Macro Criterion Pass Rate (%) | 99.3 |
| AutomationBench-AA | 77.81 | Guardrail-Adjusted Objective Completion (%) | 98.9 |
| AutomationBench-AA - Finance - Raw Objective Completion | 96.69 | Raw Objective Completion (%) | 98.9 |
| AutomationBench-AA - Strict | 53.58 | Strict Task Success (%) | 98.6 |
| AutomationBench-AA - Support - Raw Objective Completion | 93.99 | Raw Objective Completion (%) | 98.2 |
| AA GDP.pdf - Insurance - Criterion Pass Rate | 90.81 | Criterion Pass Rate (%) | 97.9 |
| AA-Briefcase - Presentation Elo | 1677.85 | Elo | 97.9 |
| AutomationBench | 53.27 | Strict Task Pass Rate (%) | 97.7 |
| AutomationBench-AA - HR - Raw Objective Completion | 81.29 | Raw Objective Completion (%) | 97.2 |
Interactive version: theaggregate.ai/model?slug=gemini-4-neon-medium · How It Works · Data refreshed daily, snapshot 2026-11-20.