Nano GPT logo
NanoGPT

Gemini 3.7 Flash

Google's frontier-performance Flash model for multimodal and agentic workloads, including coding, tool use, image understanding, PDF and document extraction, audio, and video. Google reports 65.3% on DeepSWE v1.1, up from 49.0% for Gemini 3.6 Flash, and 34% on GDP.pdf, up from 14%.

  • Reasoning
  • Vision
  • Audio Input
  • Video Input
  • Native PDF input
  • Tool Calling
  • Structured Output

Added Aug 13, 2026

Pricing

Auto routing · per 1M tokens
Input
$0.75
Output
$3.75
Cache read
$0.075
Compare provider prices

Specifications

Context window
1M
Max output
65.5K
Avg output (7d)
912 tokens
Longer than 67% of models

Benchmarks

Sourced from Artificial Analysis.

Intelligence Index

39.1

Better than 91% of models compared

Coding Index

76.1

Better than 92% of models compared

Agentic Index

35.2

Better than 74% of models compared

Agentic work

AutomationBench-AA

Workflow automation with guardrail penalties

62.0%

Better than 89% of models compared

Harvey LAB-AA

Legal agentic work criterion pass rate

90.7%

Better than 60% of models compared

AA-Briefcase

Agentic knowledge work (Elo)

1113 Elo

Better than 57% of models compared

GDPval-AA v2

Economically valuable tasks (Elo)

1392 Elo

Better than 75% of models compared

Document reasoning

GDP.pdf

Professional PDF reasoning: all-pass rate

23.6%

Better than 80% of models compared

AA-LCR v1.1

Long context reasoning with updated grading

81.7%

Better than 91% of models compared

MLCR-AA

Medical long-context reasoning

15.0%

Better than 47% of models compared

Reasoning

HLE

Humanity's Last Exam

47.9%

Better than 95% of models compared

CritPt

Research-level physics reasoning

14.3%

Coding

Terminal-Bench v4.0

Practical coding and terminal tasks

13.6%

Better than 68% of models compared

SciCode

Python programming for scientific computing

57.2%

Better than 87% of models compared

Knowledge

AA-Omniscience Accuracy

Proportion of correctly answered questions

55.3%

AA-Omniscience Hallucination Rate

Rate of incorrect answers among non-correct responses

64.5%

Legacy benchmarks

GPQA Diamond (legacy)

Graduate-level scientific reasoning

94.5%

Better than 99% of models compared

AA-LCR (unversioned / legacy)

Long context reasoning evaluation

81.7%

Better than 91% of models compared

GDPval-AA (unversioned / legacy)

Economically valuable tasks

44.6%

Last updated Oct 13, 2026

Artificial Analysis

Providers

Auto routing is available for this model. Explicit provider selection is not available.

Loading provider options…