Gemini 4 Hydrogen (Medium): 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 1899 ± 14, rank #49 of 2205 rated models, from 67 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA GDP.pdf | 32.8 | All Criteria Pass Rate (%) | 100 |
| AA GDP.pdf - Insurance - Criterion Pass Rate | 91.46 | Criterion Pass Rate (%) | 100 |
| AutomationBench-AA - Operations - Raw Objective Completion | 97.29 | Raw Objective Completion (%) | 98.8 |
| AutomationBench-AA - Finance - Raw Objective Completion | 94.25 | Raw Objective Completion (%) | 98.1 |
| AutomationBench-AA | 72.01 | Guardrail-Adjusted Objective Completion (%) | 96.5 |
| AA-Omniscience v1.1 Hallucination Rate | 12.45 | Hallucination Rate (%) | 96.4 |
| AutomationBench-AA - Strict | 43.23 | Strict Task Success (%) | 95.8 |
| AutomationBench | 46.73 | Strict Task Pass Rate (%) | 95.6 |
| AutomationBench-AA - HR - Raw Objective Completion | 78.59 | Raw Objective Completion (%) | 94.6 |
| AutomationBench-AA - Sales - Raw Objective Completion | 89.03 | Raw Objective Completion (%) | 94.6 |
| AA GDP.pdf - Engineering - Criterion Pass Rate | 81.82 | Criterion Pass Rate (%) | 93.8 |
| AA-Omniscience Index - Law | 40.22 | Omniscience Index | 92.8 |
Interactive version: theaggregate.ai/model?slug=gemini-4-hydrogen-medium · How It Works · Data refreshed daily, snapshot 2026-11-13.