STEM on XLRS-Bench macro
54.6Pass@1Seed2.0 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed2.0 Pro2026.06 | 54.6 | |
| Seed2.0 Lite2026.06 | 53.7 | |
| Gemini-3-Pro High2026.06 | 51.7 | |
| Claude-Opus-4.52026.06 | 50.4 | |
| GPT-5.2 High2026.06 | 49.9 | |
| Seed2.0 Mini2026.06 | 49.9 | |
| Seed1.82026.06 | 39.9 |