OpenAI

GPT-5.1 Codex Max

#3SWE-rebench

by OpenAI

+
+
+
+
About

GPT-5.1 Codex Max is an enhanced coding variant within the GPT-5.1 Codex line, designed for more complex software engineering tasks requiring deeper code understanding and longer agentic coding sessions. It targets professional development teams and coding agent frameworks where the standard Codex model's capabilities are insufficient for the most demanding multi-file or multi-step tasks.

+
+
+
+
Timeline
ReleasedNov 19, 2025
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown

Overall Performance

2 benchmarks
Average Score
54.4%
Best Score
60.4%
High Performers (80%+)
0

Top Categories

Coding
54.4%
+
+
+
+
All Benchmark Results for GPT-5.1 Codex Max
Complete list of benchmark scores with detailed information
Terminal Bench 2.0
Coding
60.40
60.4%
Unverified
SWE-rebench
Coding
48.50
48.5%
Unverified
+
+
+
+
Resources