GDPx: Materials Science
Science · 2026-11-01
GDPx's Materials Science domain: 23 tasks from ValenceForge and Emulated on controlling simulated lab equipment, from electron microscopes to battery cyclers, and interpreting raw experimental data. Each partner's pass@1 over three attempts per task is averaged, partners weighted equally. A component of the GDPx score, which weights its 11 domains equally.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 21.6 |
| Gemini 4 Carbon | 20.6 |
| GPT-6.1 Sol | 19.4 |
| GPT-6.1 Terra | 18.2 |
| Claude Sonnet 5.5 | 16.7 |
| Claude Opus 5.5 | 16.5 |
| Muse Wonder Full | 10.9 |
| Gemini 4 Argon | 10.2 |
| Claude Fable 5.1 | 9.7 |
| Claude Haiku 5.5 | 7.2 |
| Muse Wonder | 5.7 |
| Kimi K3 | 5.7 |
| GPT-6 Luna | 5.2 |
| Grok 4.7 | 4.4 |
| GLM-5.3 | 2.8 |
| MiniMax M3.1 Flash | 2.3 |
| Muse Spark Open | 2.0 |