reasoning benchmark
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 understanding of graph structure and edge relationships.
Updated Aug 11, 2026
Models11
Model coverage11
MetricScore
EvidenceB
Graphwalks parents <128k Ranking
Higher score ranks better on this benchmark.
| 1 | OP | 89.8% | 100.0% | 11 | C | |
| 2 | OP | 89.0% | 90.0% | 11 | C | |
| 3 | OP | 73.3% | 80.0% | 11 | C | |
| 4 | OP | 72.6% | 70.0% | 11 | C | |
| 5 | OP | 71.5% | 60.0% | 11 | C | |
| 6 | OP | 60.5% | 50.0% | 11 | C | |
| 7 | OP | 58.3% | 40.0% | 11 | C | |
| 8 | OP | 58.0% | 30.0% | 11 | C | |
| 9 | OP | 50.8% | 20.0% | 11 | C | |
| 10 | OP | 35.4% | 10.0% | 11 | C | |
| 11 | OP | 9.4% | 0.0% | 11 | C |
Graphwalks parents <128k Score Distribution
A closer view of the leading scores on this benchmark.
Graphwalks parents <128k Highlights
The leading models and scores on this benchmark.
What is Graphwalks parents <128k?
What Graphwalks parents <128k measures and how its scores work.
A graph reasoning benchmark that evaluates language models' ability to find parent nodes in graphs with context length under 128k tokens, requiring understanding of graph structure and edge relationships.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Graphwalks parents <128k
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- graphwalks-parents-<128k|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Graphwalks parents <128k.
Which model scores highest on Graphwalks parents <128k?
GPT-5.4 is currently ranked first with 89.8%.
What does Graphwalks parents <128k measure?
A graph reasoning benchmark that evaluates language models' ability to find parent nodes in graphs with context length under 128k tokens, requiring understanding of graph structure and edge relationships.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
11 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.