GPT-5.1 Codex Benchmark Details

GPT-5.1 Codex currently shows benchmark results led by Terminal-Bench (2 / 35, score 56.30), LiveCodeBench (19 / 125, score 85.50), Terminal Bench Hard (75 / 244, score 34.80).

Benchmark Results

GPT-5.1 Codex

Benchmark Results

Repository Engineering

1 evaluations

Benchmark / mode

Score

Rank/total

SWE-bench Verified

Thinking Level · HighTools

70.40

60 / 116

Algorithmic Coding

1 evaluations

Benchmark / mode

Score

Rank/total

LiveCodeBench

Thinking Level · HighTools

85.50

19 / 125

Agentic Development

2 evaluations

Benchmark / mode

Score

Rank/total

Terminal-Bench

Thinking Level · HighTools

56.30

2 / 35

Terminal Bench Hard

Thinking Level · HighTools

34.80

75 / 244

Cross-capability Suites

1 evaluations

Benchmark / mode

Score

Rank/total

LiveBench

Standard Mode

68.61

45 / 117

Visual Understanding

1 evaluations

Benchmark / mode

Score

Rank/total

MMMU-Pro

Thinking Level · High

72.50

116 / 229

Scientific Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

CritPt

Thinking Level · High

5.70

92 / 204

Code Generation & Editing

1 evaluations

Benchmark / mode

Score

Rank/total

Vibe Code Bench v1.1

Thinking Level · HighTools

13.12

54 / 61

Compare with other models