Google's fastest and most capable model with native multimodality, extreme context window, and built-in function calling for agents.
| Benchmark | Description | Score | Category | Bar |
|---|---|---|---|---|
| MMLU | Massive Multitask Language Understanding | 86.5% | knowledge | |
| HumanEval | Code generation accuracy | 89.7% | coding | |
| MATH | Mathematical reasoning | 90% | math | |
| GSM8K | Grade school math | 92% | math | |
| GPQA | Graduate-level science Q&A | 58% | reasoning | |
| HellaSwag | Common sense reasoning | 87.8% | reasoning | |
| SWE-bench | Software engineering problems | 41.1% | coding | |
| ARC-C | AI2 Reasoning Challenge | 96% | reasoning |