GraphInstruct (Generation) - L5 Multi-Step Graph Editing: leaderboard
Metric: Level quality score (0-100; the paper's 0-1 score times 100) on the 50 GraphInstruct instructions of level L5, which requires ordered edits to a base graph; zero-shot prompting; five generations per instruction at temperature 0.7 with the InstructGraph code-style prefix; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 84.1 |
| 2 | GPT-4o | 82.3 |
| 3 | Qwen 3.5 397B A17B | 82.1 |
| 4 | Qwen 3.5 122B A10B | 81.4 |
| 5 | Claude Sonnet 4 (20250514) | 80.6 |
| 6 | Llama 3.3 70B Instruct | 79.4 |
| 7 | Qwen 3.5 35B A3B | 78 |
| 8 | GPT-4.1 | 77.9 |
| 9 | GPT-3.5 Turbo | 75.2 |
| 10 | GPT-4o Mini | 75 |
| 11 | Llama 3.1 8B Instruct | 72.7 |
Interactive version: theaggregate.ai/benchmark?slug=graphinstruct-generation-l5-multi-step-graph-editing · How It Works · Data refreshed daily, snapshot 2026-10-07.