Benchmarks
347 sources · 1,258 models × 3,038 benchmarks · 158,912 scoresCondensed-matter-theory benchmark score reported in the Google Deep Think evaluation. The release gives a percentage but does not specify the exact evaluated question roster or aggregation; it remains diagnostic until those details are established.
No signal or difficulty estimate. Held out of the fit by design.
This is a benchmark-family page. Scores are recorded on the component benchmarks below rather than directly on the parent, because the components use different score scales and directions.
| Metric | Models | Leader | Direction | Date |
|---|---|---|---|---|
| CMT-Benchmark - reported pass@8 | 4 | Gemini 3 Deep Think50.5 | higher is better | 2026-02-12 |