51.6%
AA Intelligence Index v4.1
Models / Google
Gemini 3.6 · Released 2026-07-21
AA Intelligence Index 51.6 on the high-effort endpoint. Google introductory API price $0.75 / $3.75 per 1M tokens through 31 Dec 2026; AA still lists $1.50 / $7.50 on the 2026-08-13 snapshot.
Prior Google Flash workhorse, superseded for most new agent work by 3.7 Flash. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.
Family profile
Best published score in each covered benchmark family.
5 tested benchmark families
Where Gemini 3.6 Flash (high) places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| AA-LCR | Long context | #13/ 179 | 79 | Artificial Analysis | 2026-09-01 |
| AA time to first token | Performance | #13/ 186 | Artificial Analysis | 2026-09-01 | |
| AA output speed | Performance | #15/ 187 | Artificial Analysis | 2026-09-01 | |
| GPQA Diamond | Reasoning | #15/ 185 | 93 | Artificial Analysis | 2026-09-01 |
| SciCode | Coding | #29/ 180 | 53 | Artificial Analysis | 2026-09-01 |
| Humanity's Last Exam | Reasoning | #31/ 180 | 41 | Artificial Analysis | 2026-09-01 |
| Artificial Analysis Intelligence Index | Reasoning | #32/ 187 | 52 | Artificial Analysis | 2026-09-01 |
| Terminal-Bench | Agentic | #32/ 184 | 78 | Artificial Analysis | 2026-09-01 |
| τ³-Bench Banking | Agentic | #37/ 106 | 30 | Artificial Analysis | 2026-09-01 |
The three highest-scoring models with pages in each capability family. Where Gemini 3.6 Flash (high) shows up, it's highlighted.
Benchmark rows added to the public ledger for Gemini 3.6 Flash (high) in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| MMMU-Pro | Multimodal | 83.2% | Artificial Analysis — MMMU-Pro | 6 |
| AA-LCR | Long context | 79% | Artificial Analysis | 6 |
| AA output speed | Performance | 161 t/s | Artificial Analysis | 6 |
| AA time to first token | Performance | 12.64s | Artificial Analysis | 6 |
| Artificial Analysis Intelligence Index | Reasoning | 51.6% | Artificial Analysis | 6 |
| GPQA Diamond | Reasoning | 92.8% | Artificial Analysis | 6 |
| Humanity's Last Exam | Reasoning | 40.8% | Artificial Analysis | 6 |
| SciCode | Coding | 52.7% | Artificial Analysis | 6 |
| τ³-Bench Banking | Agentic | 29.9% | Artificial Analysis | 6 |
| Terminal-Bench | Agentic | 77.53% | Artificial Analysis | 6 |
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
Every catalog benchmark for Gemini 3.6 Flash (high). Scores link to the original source; gaps mean no public row exists yet.
10 of 37 catalog benchmarks have a sourced row for Gemini 3.6 Flash (high).
51.6%
AA Intelligence Index v4.1
92.8%
GPQA Diamond (AA run)
40.8%
HLE (AA run)
52.7%
SciCode (AA run)
77.53%
Terminal-Bench v2.1 (AA run)
29.9%
τ³-Bench Banking (AA run)
83.2%
MMMU-Pro (AA run)
79%
AA-LCR (AA run)
161 t/s
Median output tokens/s (1k prompt, default provider)
12.64s
Median time to first token (1k prompt, default provider)