Atlas

Benchmarks

← All benchmarks

GraphWalks BFS: Up to 128K

General QA · 2025-04-12

Mean node-set F1 across Google's aggregate of 650 GraphWalks BFS problems with context lengths up to 128K tokens. This aggregate is not an individual context-length bin.

Top models (higher is better)

ModelScore
Gemini 4 Argon99.7
GPT-6 Astra98.7
Claude Fable 5.191.4
Claude Opus 5.590.6
Loading Atlas data…