First Flash model with thinking, near-Pro performance at one-eighth the price
| Benchmark | Score | Rank |
|---|---|---|
SWE-bench Real-world GitHub issue resolution | 60.4% | #30 / 39 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 80.8% | #30 / 59 |
MMLU Tests knowledge across 57 subjects from STEM to humanities | 88.4% | #32 / 54 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 76.2% | #47 / 59 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 83.7% | #48 / 58 |
TerminalArtificial Analysis Agentic terminal coding tasks requiring multi-step execution | 16.7% | #49 / 56 |
GPQA PhD-level science questions even experts struggle with | 82.8% | #64 / 91 |
hleArtificial Analysis | 12.7% | #64 / 79 |