Benchmark
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…
- Modality
- text
- Categories
- long_context, reasoning, spatial_reasoning
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 7
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Claude Opus 4.6 | Anthropic | 0.954 | 2026-02-05 | evidence |
| Claude Opus 4.8 | Anthropic | 0.833 | 2026-05-28 | evidence |
| GPT-4.1 | OpenAI | 0.25 | 2025-04-14 | evidence |
| GPT-4.1 mini | OpenAI | 0.11 | 2025-04-14 | evidence |
| GPT-4.1 nano | OpenAI | 0.056 | 2025-04-14 | evidence |
| GPT-5.4 | OpenAI | 0.324 | 2026-03-05 | evidence |
| GPT-5.5 | OpenAI | 0.585 | 2026-04-23 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.