GraphInstruct (Generation) - L1 Single Constraint: leaderboard
Metric: Level quality score (0-100; the paper's 0-1 score times 100) on the 200 GraphInstruct instructions of level L1, which requires a single explicit constraint (graph type); 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: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4 (20250514) | 98.3 |
| 2 | GPT-4o | 98.2 |
| 3 | Qwen 3.5 122B A10B | 98 |
| 4 | Qwen 3.5 397B A17B | 97.9 |
| 5 | Claude Sonnet 4.6 | 97.7 |
| 6 | GPT-4.1 | 96.8 |
| 7 | Qwen 3.5 35B A3B | 95.5 |
| 8 | Llama 3.3 70B Instruct | 94.8 |
| 9 | GPT-3.5 Turbo | 90.5 |
| 10 | GPT-4o Mini | 89.8 |
| 11 | Llama 3.1 8B Instruct | 73 |
Interactive version: theaggregate.ai/benchmark?slug=graphinstruct-generation-l1-single-constraint · How It Works · Data refreshed daily, snapshot 2026-10-07.