# GraphWalks BFS 256K–1M

> Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens.

Canonical page: https://benchlm.ai/benchmarks/graphwalks-bfs-256k-1m

- Category: [Reasoning](/reasoning)
- Last updated: October 14, 2026

## About GraphWalks BFS 256K–1M

- Year: 2026
- Tasks: Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens.
- Format: F1
- Difficulty: Frontier agent evaluation
- Paper: [Gemini 4 Argon evaluation methodology](https://deepmind.google/models/evals-methodology/gemini-4-argon)

Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens. We keep this versioned result as display-only evidence.

GraphWalks BFS 256K–1M is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (4 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Gemini 4 Argon](/models/gemini-4-argon) | Google | 84.2% |
| 2 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | 71.8% |
| 3 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | 66.8% |
| 4 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | 65.0% |

## FAQ

### What does GraphWalks BFS 256K–1M measure?

Breadth-first-search graph traversal on 200 problems with context lengths from 256K to 1M tokens.

### Which model scores highest on GraphWalks BFS 256K–1M?

Gemini 4 Argon by Google currently leads with a score of 84.2% on GraphWalks BFS 256K–1M.

### How many models are evaluated on GraphWalks BFS 256K–1M?

4 AI models have published results on GraphWalks BFS 256K–1M in the BenchLM catalog.

### Does GraphWalks BFS 256K–1M affect BenchLM's overall score?

Not directly. GraphWalks BFS 256K–1M is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on GraphWalks BFS 256K–1M

- [Gemini 4 Argon vs GPT-6 Astra](/compare/gemini-4-argon-vs-gpt-6-astra)
- [GPT-6 Astra vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-gpt-6-astra)
- [Claude Opus 5.5 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-opus-5-5)