Benchmark
Graphwalks BFS >128k Benchmark
What is Graphwalks BFS >128k?
A graph reasoning benchmark that evaluates language models' ability to perform breadth-first search (BFS) operations on graphs with context length over 128k…
- Modality
- text
- Evaluates
- long context, reasoning, spatial reasoning
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 11
Source provenance
Graphwalks BFS >128k results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Claude Mythos Preview | Anthropic | 0.8 | 2026-04-07 | evidence |
| Claude Opus 4.6 | Anthropic | 0.615 | 2026-02-05 | evidence |
| Claude Opus 4.8 | Anthropic | 0.681 | 2026-05-28 | evidence |
| GPT-4.1 | OpenAI | 0.19 | 2025-04-14 | evidence |
| GPT-4.1 mini | OpenAI | 0.15 | 2025-04-14 | evidence |
| GPT-4.1 nano | OpenAI | 0.029 | 2025-04-14 | evidence |
| GPT-5.4 | OpenAI | 0.214 | 2026-03-05 | evidence |
| GPT-5.5 | OpenAI | 0.454 | 2026-04-23 | evidence |
| GPT-5.6 Luna | OpenAI | 0.813 | 2026-07-09 | evidence |
| GPT-5.6 Sol | OpenAI | 0.907 | 2026-07-09 | evidence |
| GPT-5.6 Terra | OpenAI | 0.769 | 2026-07-09 | 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.