Gemini 4 Hydrogen (Low): benchmark results
Google DeepMind model adapted from Gemini 4 Carbon, released publicly through the Gemini API and apps on November 12, 2026. Low, medium, high (default) and xhigh thinking levels; text, image, audio and video inputs. Provider: Google. Released 2026-11-12. Access: API.
Aggregate Rating 1840 ± 17, rank #95 of 2205 rated models, from 66 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA GDP.pdf - Insurance - Criterion Pass Rate | 88.54 | Criterion Pass Rate (%) | 96.3 |
| AA GDP.pdf - STEM/Research - Criterion Pass Rate | 79.34 | Criterion Pass Rate (%) | 93.6 |
| AA GDP.pdf | 28.2 | All Criteria Pass Rate (%) | 91.9 |
| AA-Omniscience v1.1 Hallucination Rate | 18.85 | Hallucination Rate (%) | 91.4 |
| AA-Omniscience Index - Science, Engineering & Mathematics | 35.17 | Omniscience Index | 90.6 |
| AutomationBench-AA - Finance - Raw Objective Completion | 88.93 | Raw Objective Completion (%) | 90 |
| AA-Omniscience Index - Law | 35.33 | Omniscience Index | 88.5 |
| AA GDP.pdf - Real Estate - Criterion Pass Rate | 73.43 | Criterion Pass Rate (%) | 88 |
| AutomationBench | 30.59 | Strict Task Pass Rate (%) | 87.7 |
| AA GDP.pdf - Engineering - Criterion Pass Rate | 78.62 | Criterion Pass Rate (%) | 87.6 |
| AutomationBench-AA - Marketing - Raw Objective Completion | 86.51 | Raw Objective Completion (%) | 86.2 |
| AA GDP.pdf - Criterion Pass Rate | 78.2 | Macro Criterion Pass Rate (%) | 83.7 |
Interactive version: theaggregate.ai/model?slug=gemini-4-hydrogen-low · How It Works · Data refreshed daily, snapshot 2026-11-13.