Model

Gemini 3.7 Flash

Release Date: Aug 13, 2026

Developer Google 🇺🇸

Context Window 1M

Max Output Tokens 66k

Token Costs (in/out) $1.50/7.50

Weights Private

Input Modalities

Accuracy

59.31 % ± 1.06

Cost / Test (Vals Index)

$ 4.170

Latency

43 min 13 s

Vals Index

BenchmarksAccuracyRankings

59.31%

±1.06

19/66

18.27%

17/22

34.80%

±4.22

31/69

CyberBench v1.1

43.75%

±2.21

37/41

71.33%

±2.26

10/66

Finance Agent (v2)

59.04%

±0.27

4/70

Legal Research Bench

34.62%

±3.31

33/69

53.39%

±2.12

8/102

83.94%

±2.00

36/104

66.65%

±0.92

33/98

ProofBench v1.1

58.00%

±4.96

17/41

49.23%

±3.38

22/88

4.58%

±1.29

7/14

Tax Agent Bench

57.66%

±3.31

34/61

74.73%

±0.85

31/145

Vibe Code Bench v1.1

70.39%

±4.84

31/104

Harvey's Legal Agent Benchmark

8.75%

±2.15

17/70

93.94%

±1.49

5/138

67.83%

±3.61

11/36

88.65%

±0.92

5/143

87.26%

±0.42

4/148

90.12%

±0.30

6/138

88.96%

±0.75

5/93

0.00%

±0.00

26/53

65.89%

±4.55

3/35

80.80%

±1.76

23/88

Terminal-Bench 4.0

6.06%

±0.88

27/37

Terminal-Bench Science

5.71%

±2.79

16/33

Proprietary BenchmarksAcademic BenchmarksIndustry Partners

Vals

View Hyperparameter Settings

Default Provider : Google

Temperature: 1

Top P: Default

Top K: Default

Max Output Tokens: 65,536

Reasoning Effort: high

Compared with

Gemini 3.7 Flash vs Grok 4.6 25 shared benchmarks Gemini 3.7 Flash vs MiMo V2.6 Pro 18 shared benchmarks Gemini 3.7 Flash vs MiMo V2.6 Flash 18 shared benchmarks Gemini 3.7 Flash vs Claude Sonnet 5 26 shared benchmarks Gemini 3.7 Flash vs GPT-5.6 Terra 26 shared benchmarks Gemini 3.7 Flash vs GPT-5.6 Luna 25 shared benchmarks Gemini 3.7 Flash vs GPT-6 Luna 17 shared benchmarks Gemini 3.7 Flash vs Grok 4.7 19 shared benchmarks Gemini 3.7 Flash vs GPT 5.5 23 shared benchmarks Gemini 3.7 Flash vs Kimi K3 26 shared benchmarks