Fast Gemini workhorse for multimodal apps where latency and price matter
Benchmarks
| Benchmark | Category | Score | Variant | Provenance | Source | Observed |
|---|---|---|---|---|---|---|
| AA Intelligence Index | intelligence | 20.1 | thinking | independent | aa | Jul 24, 2026 |
| Design Arena — 3d | preference | 1131 | — | mirror | openrouter | Jul 23, 2026 |
| Design Arena — codecategories | preference | 1137 | — | mirror | openrouter | Jul 23, 2026 |
| Design Arena — dataviz | preference | 1161 | — | mirror | openrouter | Jul 24, 2026 |
| Design Arena — gamedev | preference | 1124 | — | mirror | openrouter | Jul 23, 2026 |
| Design Arena — svg | preference | 1069 | — | mirror | openrouter | Jul 24, 2026 |
| Design Arena — uicomponent | preference | 1133 | — | mirror | openrouter | Jul 24, 2026 |
| Design Arena — website | preference | 1143 | — | mirror | openrouter | Jul 24, 2026 |
| LMArena Text | preference | 1465 | — | crowd | lmarena | Jul 21, 2026 |
| LMArena Vision | preference | 1236 | — | crowd | lmarena | Jul 21, 2026 |
| Median output speed | speed | 231 tok/s | thinking | independent | aa | Jul 24, 2026 |
| Median time to first token | speed | 17.58s | thinking | independent | aa | Jul 24, 2026 |
Includes index data from Artificial Analysis.
Providers
Live operational stats per hosting endpoint, via OpenRouter.
| Provider | Quant | Context | $/1M in · out | Uptime 24h | Latency | Throughput |
|---|---|---|---|---|---|---|
| Google AI Studio | — | 1.0M | $0.3 · $2.5 | 99.95% | — | — |
| Google AI Studio | — | 1.0M | $0.15 · $1.25 | 99.95% | — | — |
| Google AI Studio | — | 1.0M | $0.54 · $4.5 | 99.95% | — | — |
| — | 1.0M | $0.3 · $2.5 | 86.84% | — | — | |
| — | 1.0M | $0.3 · $2.5 | 96.97% | — | — | |
| — | 1.0M | $0.3 · $2.5 | 98.25% | — | — | |
| — | 1.0M | $0.54 · $4.5 | 98.25% | — | — |