Gemini 4 Argon Benchmark Details

Gemini 4 Argon currently shows benchmark results led by DeepSWE (1 / 92, score 77.90), Vibe Code Bench v1.1 (1 / 61, score 91.90), Finance Agent v2 (1 / 43, score 65.40).

Benchmark Results

Gemini 4 Argon

Benchmark Results

Repository Engineering

1 evaluations

Benchmark / mode

Score

Rank/total

DeepSWE

Thinking Level · MaxTools

77.90

1 / 92

Desktop Workflows

1 evaluations

Benchmark / mode

Score

Rank/total

OSWorld 2.0

Thinking Level · MaxTools

69.20

5 / 14

Office & Business

1 evaluations

Benchmark / mode

Score

Rank/total

AutomationBench

Thinking Level · MaxTools

51.30

4 / 25

Finance

1 evaluations

Benchmark / mode

Score

Rank/total

Finance Agent v2

Thinking Level · MaxTools

65.40

1 / 43

Video Understanding

1 evaluations

Benchmark / mode

Score

Rank/total

LVBench

Thinking Level · Max

91.70

1 / 5

Biology & Genomics

1 evaluations

Benchmark / mode

Score

Rank/total

LABBench2

Thinking Level · MaxToolsInternet

88.80

1 / 3

Documents & Charts

1 evaluations

Benchmark / mode

Score

Rank/total

Chartography

Thinking Level · Max

71.60

7 / 11

Agentic Development

1 evaluations

Benchmark / mode

Score

Rank/total

Terminal-Bench 4.0

Thinking Level · MaxTools

57.40

9 / 98

Code Generation & Editing

1 evaluations

Benchmark / mode

Score

Rank/total

Vibe Code Bench v1.1

Thinking Level · MaxTools

91.90

1 / 61

Capability Indices

1 evaluations

Benchmark / mode

Score

Rank/total

Vals Index v2.1

Thinking Level · HighTools

68.90

1 / 36

Compare with other models