Nanbeige

Nanbeige · 1 models

Nanbeige's compact LLM lab uses Looped Transformer architecture to maximize capacity per parameter. Nanbeige4.2-3B outperforms 9B+ models on agentic and reasoning benchmarks while fitting in under 2.5 GB at Q4.

Models in this family

1 models in the Nanbeige family, grouped by series. Q4 (GB) is weights-only; total VRAM adds KV cache and overhead — the fit check links compute it for a representative retail GPU at 8K context.

Nanbeige 4.2