GPT-5.5
Open paper
Benchmark
Model
Rank
Results
code-generation-on-terminal-bench-2-1
Codex + GPT-5.5
–
Accuracy: 83.1
code-generation-on-terminal-bench-2-1
Terminus 2 + GPT-5.5
–
Accuracy: 78.0
Rank counts only results with a code link.