Which AI model scores highest on GraphWalks BFS (<128K)?

GPT-5.2 by OpenAI holds the best GraphWalks BFS (<128K) result among tracked models, at 94% (released Dec 11 2025). Higher scores are better on this benchmark.

What are the top 5 models on GraphWalks BFS (<128K)?

1. GPT-5.2 (OpenAI) — 94%; 2. GPT-5.4 (OpenAI) — 93%; 3. GPT-5 (OpenAI) — 78.3%; 4. GPT-5.4 mini (OpenAI) — 76.3%; 5. GPT-5 mini (OpenAI) — 73.4%.

How many models have a published GraphWalks BFS (<128K) score?

10 tracked models have a published GraphWalks BFS (<128K) 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.

Long context

GraphWalks BFS (<128K)

breadth-first traversal accuracy for inputs shorter than 128K tokens

Rankings

Higher is better

GraphWalks BFS (<128K) — frequently asked questions

: breadth-first traversal accuracy for inputs shorter than 128K tokens
: GPT-5.2 by OpenAI holds the best GraphWalks BFS (<128K) result among tracked models, at 94% (released Dec 11 2025). Higher scores are better on this benchmark.
: 1. GPT-5.2 (OpenAI) — 94%; 2. GPT-5.4 (OpenAI) — 93%; 3. GPT-5 (OpenAI) — 78.3%; 4. GPT-5.4 mini (OpenAI) — 76.3%; 5. GPT-5 mini (OpenAI) — 73.4%.
: 10 tracked models have a published GraphWalks BFS (<128K) 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.