Which AI model scores highest on Expert-SWE (Internal)?

GPT-5.5 by OpenAI holds the best Expert-SWE (Internal) result among tracked models, at 73.1% (released Apr 23 2026). Higher scores are better on this benchmark.

What are the top 2 models on Expert-SWE (Internal)?

1. GPT-5.5 (OpenAI) — 73.1%; 2. GPT-5.4 (OpenAI) — 68.5%.

How many models have a published Expert-SWE (Internal) score?

2 tracked models have a published Expert-SWE (Internal) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

Software engineering

Expert-SWE (Internal)

OpenAI's private set of expert-level software-engineering problems. Higher is better.

Rankings

Higher is better

Expert-SWE (Internal) — frequently asked questions

: OpenAI's private set of expert-level software-engineering problems. Higher is better.
: GPT-5.5 by OpenAI holds the best Expert-SWE (Internal) result among tracked models, at 73.1% (released Apr 23 2026). Higher scores are better on this benchmark.
: 1. GPT-5.5 (OpenAI) — 73.1%; 2. GPT-5.4 (OpenAI) — 68.5%.
: 2 tracked models have a published Expert-SWE (Internal) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.