Nanbeige 4.2 (3B, Q4)
Nanbeige LLM Lab's compact agentic reasoning model. A looped transformer — 22 layers run twice over shared weights — so a 3B-parameter download does the compute of a much deeper stack. Text-only, with native thinking, XML tool calling, English + Chinese, and a 256K context window. Q4 builds suit machines with 8GB+ usable memory.
| At a glance | |
|---|---|
| Made by | Nanbeige LLM Lab |
| Download size | 2.7 GB |
| Parameters | 3B |
| Quantization | Q4_K_M |
| Context window | 262K tokens |
| Uses tools | Yes |
| Best for | Reasoning |
| Runs on | llama.cpp |
| License | Apache 2.0 |
| Added to the catalog | 2026-08-31 |
Gezel downloads this model the first time a gezel needs it, checks the file against its published hash, and applies tuned settings for it.