Across 14 shared benchmarks, Agnes 2.5 Pro Beta scores higher on 13 and MiniCPM5-2B on 1. The widest gap is Terminal-Bench 2.1, where Agnes 2.5 Pro Beta scores 69.7 against 8.6.
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.