all the models — AI benchmark observatory
← Benchmarks

Graphwalks parents >128k

A graph reasoning benchmark that evaluates language models' ability to find parent nodes in graphs with context length over 128k tokens, testing long-context reasoning and graph structure understanding.

id graphwalks-parents->128k · max 1 · 7 models reported

#ModelScore
1Claude Opus 4.6
Anthropic
0.95
Graphwalks parents >128k Leaderboard · all the models