ChemGraph - Single-Agent: leaderboard
ChemGraph Leaderboard (Argonne): 40 computational-chemistry queries in 12 categories, including SMILES lookup, geometry optimization, vibrational frequencies, thermochemistry, dipole moments and reaction Gibbs energies, solved with ASE-based tools by a single agent. A structured judge accepts answers within 5% relative tolerance; accuracy %, higher is better.
Metric: Accuracy (%, 40 computational-chemistry queries). Source: huggingface.co. Saturation forecast: Estimated already saturated. 32 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 100 | #148 |
| 2 | GPT-5.5 | 100 | #33 |
| 3 | Claude Haiku 4.5 | 100 | #287 |
| 4 | Claude Opus 4.7 | 100 | #51 |
| 5 | GPT-5.6 Sol | 100 | #22 |
| 6 | Claude Opus 4.8 | 100 | #39 |
| 7 | Gemini 3.5 Flash | 100 | #66 |
| 8 | O3 | 100 | #132 |
| 9 | Claude Sonnet 5 | 100 | #65 |
| 10 | Nemotron 3 Ultra | 100 | #245 |
| 11 | Gemini 2.5 Pro | 97.5 | #159 |
| 12 | Claude Sonnet 4.6 | 97.5 | #94 |
| 13 | O4 Mini | 97.5 | #185 |
| 14 | GPT-4o | 95 | #357 |
| 15 | GPT-5.6 Luna | 95 | #79 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=chemgraph-single-agent · How It Works · Data refreshed daily, snapshot 2026-11-05.