GraphInstruct (Generation) - L4 Domain Semantics: leaderboard
GraphInstruct (Generation) L4 Domain Semantics scores zero-shot graph generation on the 100 GraphInstruct (2026) instructions whose graphs must satisfy domain semantics (social, citation, molecular graphs). The level quality score, 0 to 100 and higher is better, is computed by code from validity, distribution match, reference similarity and constraint satisfaction.
Metric: Level quality score (0-100; the paper's 0-1 score times 100) on the 100 GraphInstruct instructions of level L4, which requires domain semantics (social, citation, molecular graphs); 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: Not forecast. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 76.3 | #173 |
| 2 | GPT-4o | 75.1 | #751 |
| 3 | Llama 3.3 70B Instruct | 74.6 | #1385 |
| 4 | GPT-4o Mini | 74.4 | #1667 |
| 5 | GraphInstruct DeepSeek-V3 (checkpoint unspecified) | 74.4 | |
| 6 | GPT-3.5 Turbo | 74.2 | #3005 |
| 7 | GPT-4.1 | 73.7 | #505 |
| 8 | Claude Sonnet 4 (20250514) | 73.7 | #405 |
| 9 | Qwen 3.5 122B A10B | 72.4 | #346 |
| 10 | Qwen 3.5 397B A17B | 72.1 | #264 |
| 11 | Qwen 3.5 35B A3B | 72.1 | #492 |
| 12 | Llama 3.1 8B Instruct | 70.3 | #4600 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=graphinstruct-generation-l4-domain-semantics · How It Works · Data refreshed daily, snapshot 2026-11-21.