Gemini 3.7 Flash
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Multimodal#1087.2
- 2Reasoning81.6
- 3Knowledge#1780.7
- 4Coding#2578.6
- 5Math74.8
- 6Agentic & tool use#3164.7
Benchmark results (24)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| Terminal-Bench 2.1 | 85.8 | #15 of 68 |
| MazeBench | 1 | #13 of 42 |
| AutomationBench | 30.4 | #21 of 25 |
| OSWorld 2.0 | 47.9 | #11 of 24 |
| ApprenticeBench | 16 | #10 of 18 |
| Agents' Last Exam | 26.3 | #13 of 17 |
| Terminal-Bench 3.0 | 14.9 | #8 of 13 |
Multimodal
| Design Arena Website | 1304 | #13 of 91 |
| CharXiv | 88.7 | #10 of 42 |
| ParseBench | 71.27 | #7 of 37 |
| CharXiv w/o tools | 84.5 | #6 of 14 |
| LVBench | 85.4 | #3 of 6 |
Knowledge
| LABBench2 | 82.1 | #4 of 7 |
| HLE-Verified | 53.6 | #4 of 6 |
| BioMysteryBench (human-difficult) | 43.5 | #5 of 6 |
| BioMysteryBench (human-solvable) | 87.1 | #5 of 5 |
Coding
| DeepSWE | 65.3 | #30 of 50 |
| FrontierSWE v2 | 20.3 | #15 of 20 |
| FrontierCode 1.1 Main | 43.6 | #9 of 16 |
Reasoning
| ARC-AGI-2 | 84.6 | #12 of 41 |
| ARC-AGI-1 | 95.50 | #11 of 36 |
| MRCR v2 64K-128K | 97.0 | #1 of 2 |
Math
| FrontierMath v2 (Tiers 1-3) | 71.579 | #22 of 64 |
| FrontierMath v2 (Tier 4) | 36.585 | #22 of 51 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price (BenchLM) | $0.75 | $3.75 | $0.075 |
| OpenRouter google/gemini-3.7-flash | $0.75 | $3.75 | $0.075 |
Google lists introductory Gemini 3.7 Flash rates through December 31, 2026: $0.75 input, $0.075 cached input, and $3.75 output per million tokens. Flex and Batch input and output rates are half those amounts. On January 1, 2027, standard rates rise to $1.50 input, $0.15 cached input, and $7.50 output; Flex and Batch rates rise to $0.75 input, $0.075 cached input, and $3.75 output. Priority rates are $1.35 input, $0.135 cached input, and $6.75 output during the promotion, doubling in 2027.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text(High) | 1487 | #14 · 27.2K votes |
| WebDev (coding)(High) | 1591 | #32 · 12.6K votes |
| Vision(High) | 1314 | #8 · 5.4K votes |
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
- Gemini 3.6 Flash62.8
- Gemini 3.5 Flash61.9
- Gemini 3 Flash54.3
Similar score
- Grok 4.667.1 · $3.00
- MiniMax M3.1 Flash67.1 · $0.26
- GPT-5.466.7 · $5.63
- GPT-5.2 Pro66.7 · $57.75
- Claude Haiku 5.567.2 · $0.20