Skip to content

GraphWalks BFS 256K–1M

Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens.

4 models

Higher is better.

  1. 1Gemini 4 ArgonGoogle84.2
  2. 2GPT-6 AstraOpenAI71.8
  3. 3Claude Opus 5.5Anthropic66.8
  4. 4Claude Fable 5.1Anthropic65.0
Higher is better. Bar colour: blue proprietary, orange open weights, green other.

About this benchmark

Metric
F1
Tasks
Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens.
Authors
Google DeepMind (2026)

Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.