A compact, efficient text model that punches above its weight class for a 3-billion parameter system. It handles a surprisingly long context window of 128K tokens, making it capable of processing lengthy documents despite its small footprint. The trade-off is typical of small models — it won't match larger systems on complex reasoning, but it runs lean and is freely available for self-hosting.
| Benchmark | Score | Type | Recorded |
|---|---|---|---|
| IFEval | 13.6 | accuracy | 3d ago |