Quality
Highest overall scores
- 1
Google: Gemini 3.5 Flash52 - 2Google: Gemini 3.6 Flash51.6
- 3Google: Gemini 3.1 Pro Preview46.5
Google model benchmarks.
Compare Google models across answer quality, coding, complex tasks, speed, price, and context size.
Response speed
Fastest output in this sample
- 1Google: Gemini 3.5 Flash231.5 t/s
- 2Google: Gemini 3.6 Flash202 t/s
- 3Google: Gemini 3.1 Pro Preview101 t/s
Input price
Lowest price per 1M tokens
- 1Google: Gemini 3.5 Flash$1.50/1M
- 2Google: Gemini 3.6 Flash$1.50/1M
- 3Google: Gemini 3.1 Pro Preview$2.00/1M
Highlights
Quality
Overall capability · Higher is better
Speed
Output tokens per second · Higher is better
Input price
USD per 1M tokens · Lower is better
Quality benchmarks
Compare Google model capability.
Performance and price
Compare practical tradeoffs.
All Google models.
3 models in the current catalogue.
| Model | Links | |||||
|---|---|---|---|---|---|---|
| Google: Gemini 3.5 Flash | 52 | 231.5 t/s | 1M | $1.50 | $9.00 | |
| Google: Gemini 3.6 Flash | 51.6 | 202 t/s | 1M | $1.50 | $7.50 | |
| Google: Gemini 3.1 Pro Preview | 46.5 | 101 t/s | 1M | $2.00 | $12.00 |
Google technical benchmarks.
Compare named evaluation records available for Google models. Missing results remain blank rather than being estimated.
| Benchmark | Area | Gemini 3.5 Flash | Gemini 3.6 Flash | Gemini 3.1 Pro Preview | Source |
|---|---|---|---|---|---|
| Intelligence Index | overallQuality | 52 | 51.6 | 46.5 | OpenRouter / artificial-analysis |
| Coding Index | coding | 70.1 | 69.2 | 68.8 | OpenRouter / artificial-analysis |
| Agentic Index | agentic | 39.7 | 40.5 | 21.4 | OpenRouter / artificial-analysis |
| GPQA | GPQA Diamond | 92.2 | 92.8 | 94.1 | OpenRouter model benchmarks |
| Humanity's Last Exam | HLE | 42.7 | 40.8 | 44.7 | OpenRouter model benchmarks |
| IFBench | IFBench | 76.3 | — | 77.1 | OpenRouter model benchmarks |
| τ²-Bench Telecom | τ²-Bench Telecom | 95.3 | — | 95.6 | OpenRouter model benchmarks |
| AA-LCR | AA-LCR | 81 | 79 | 72.7 | OpenRouter model benchmarks |
| GDPval-AA | GDPval-AA | 42.2 | 46.1 | 23.1 | OpenRouter model benchmarks |
| CritPt | CritPt | 13.1 | 10.6 | 17.7 | OpenRouter model benchmarks |
| SciCode | SciCode | 53.1 | 52.7 | 58.9 | OpenRouter model benchmarks |
| Terminal-Bench Hard | Terminal-Bench Hard | 40.9 | — | 53.8 | OpenRouter model benchmarks |
| AA-Omniscience Accuracy | AA-Omniscience Accuracy | 51.4 | 50 | 55.2 | OpenRouter model benchmarks |
| AA-Omniscience Non-Hallucination Rate | AA-Omniscience Non-Hallucination Rate | 37.8 | 44.4 | 50.1 | OpenRouter model benchmarks |
Google image models.
| Model | Quality | Evidence | Price /1k images | Links |
|---|---|---|---|---|
| Nano Banana 2 (Gemini 3.1 Flash Image Preview) | 1264 Elo | 19,895 samples | $67 | View details |
| Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 1263 Elo | 10,121 samples | $33.6 | View details |
| Nano Banana Pro (Gemini 3 Pro Image) | 1226 Elo | 12,280 samples | $134 | View details |
| Nano Banana (Gemini 2.5 Flash Image) | 1153 Elo | 11,198 samples | $39 | View details |
| Nano Banana 2 (Gemini 3.1 Flash Image) | — | — | — | View details |
| Nano Banana Pro (Gemini 3 Pro Image Preview) | — | — | — | View details |
Google video models.
| Model | Quality | Text to video | Image to video | Links |
|---|---|---|---|---|
| Veo 3.1 | 68 | 1099 Elo | 1086 Elo | View details |
| Veo 3.1 Fast | 58 | 1092 Elo | 1076 Elo | View details |
| Veo 3.1 Lite | 46 | 1090 Elo | 1064 Elo | View details |