Benchmark
Graphwalks parents <128k Benchmark
What is Graphwalks parents <128k?
A graph reasoning benchmark that evaluates language models' ability to find parent nodes in graphs with context length under 128k tokens, requiring…
- Modality
- text
- Evaluates
- reasoning, spatial reasoning
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 11
Source provenance
Graphwalks parents <128k results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| GPT-4.1 | OpenAI | 0.58 | 2025-04-14 | evidence |
| GPT-4.1 mini | OpenAI | 0.605 | 2025-04-14 | evidence |
| GPT-4.1 nano | OpenAI | 0.094 | 2025-04-14 | evidence |
| GPT-4.5 | OpenAI | 0.726 | 2025-02-27 | evidence |
| GPT-4o | OpenAI | 0.354 | 2024-08-06 | evidence |
| GPT-5 | OpenAI | 0.733 | 2025-08-07 | evidence |
| GPT-5.2 | OpenAI | 0.89 | 2025-12-11 | evidence |
| GPT-5.4 | OpenAI | 0.898 | 2026-03-05 | evidence |
| GPT-5.4 mini | OpenAI | 0.715 | 2026-03-17 | evidence |
| GPT-5.4 nano | OpenAI | 0.508 | 2026-03-17 | evidence |
| o3-mini | OpenAI | 0.583 | 2025-01-30 | 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.