High-throughput Flash model for scaled agentic search and document processing, with 350 output tokens per second reported by Artificial Analysis.
SI v3 uses weighted percentages from one Artificial Analysis configuration. Ranking requires all five benchmarks. Partial scores average only available weights. Scores are not comparable to SI v2 or a probability of reaching singularity.
Compared with models released within six months of it.
| Benchmark | Score | Rank |
|---|---|---|
aaAgentic | 26.8% | #16 / 16 |
AA Intelligence (Historical) Archived release-era composite, not comparable to the current versioned Intelligence Index used for rankings | 36.5% | #22 / 22 |
AA Coding (Historical) Archived release-era coding composite, excluded from current rankings because comparable current-series coverage is unavailable | 49.3% | #22 / 22 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 83.6% | #29 / 51 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 85.8% | #42 / 61 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 79% | #47 / 62 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 83.8% | #65 / 96 |
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 18.8% | #82 / 102 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisGemini 3.5 Flash-Lite | 18.8% |
MMMU-Pro Artificial AnalysisGemini 3.5 Flash-Lite | 79% |
Terminal-Bench 2.1 Artificial AnalysisGemini 3.5 Flash-Lite | 53.6% |
Terminal-Bench 4.0 Artificial AnalysisGemini 3.5 Flash-Lite | 1% |
SciCode Artificial AnalysisGemini 3.5 Flash-Lite | 41.3% |
AA-LCR Artificial AnalysisGemini 3.5 Flash-Lite | 76% |
CritPt Artificial AnalysisGemini 3.5 Flash-Lite | 0% |