Compare Models
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
Vals Index

0.0%

±1.06
13/58
Code Migration

0.0%

±4.22
24/61
EMB

0.0%

±2.26
7/58
Finance Agent (v2)

0.0%

±0.27
4/61
Legal Research Bench

0.0%

±3.31
28/61
MedCode

0.0%

±2.12
8/93
MedScribe

0.0%

±2.00
31/95
MortgageTax

0.0%

±0.92
33/98
ProofBench v1.1

0.0%

±4.96
11/32
SAGE

0.0%

±3.38
21/81
SRE Bench

0.0%

±1.29
5/10
Tax Agent Bench

0.0%

±3.31
18/21
TaxEval v2

0.0%

±0.85
31/145
Vibe Code Bench v1.1

0.0%

±4.84
24/96
Harvey's Legal Agent Benchmark

0.0%

±2.15
14/62
GPQA Diamond

0.0%

±1.49
5/138
IOI

0.0%

±3.61
8/28
LiveCodeBench

0.0%

±0.92
5/143
LegalBench

0.0%

±0.42
4/145
MMLU Pro

0.0%

±0.30
6/138
MMMU Pro

0.0%

±0.75
5/93
ProgramBench

0.0%

±0.00
20/47
SkillsBench

0.0%

±4.55
3/35
SWE-bench

0.0%

±1.76
23/88
Terminal-Bench 4.0

0.0%

±0.88
20/28
Terminal-Bench Science

0.0%

±2.79
11/26
Proprietary BenchmarksAcademic BenchmarksIndustry Partners
Vals
Default Provider : Google
Temperature: 1
Top P: Default
Top K: Default
Max Output Tokens: 65,536
Reasoning Effort: high