Successor to 3.5 Flash, pushing Flash-tier reasoning to GPQA 92.8 while holding the cost-efficient position.
| Benchmark | Score | Rank |
|---|---|---|
aaAgentic | 38.7% | #3 / 5 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 88.1% | #4 / 53 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 88.4% | #4 / 51 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 89.3% | #8 / 51 |
AA Intelligence Independent composite of agentic work, coding, scientific reasoning, and general capability | 50.1% | #10 / 11 |
AA Coding Independent composite of implementation, terminal use, and real-codebase agent performance | 69.2% | #10 / 11 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 92.8% | #11 / 78 |
hleArtificial Analysis | 38.3% | #18 / 66 |