A compact specialist that punches above its weight in mathematical reasoning. Despite its small footprint, it handles arithmetic, algebra, and structured problem-solving with focused precision — the result of math-specific training rather than general-purpose breadth. Outside of quantitative tasks, its narrow specialization shows quickly.
| Benchmark | Score | Type | Recorded |
|---|---|---|---|
| IFEval | 18.6 | accuracy | 4d ago |