Sakura Nanbeige4.2 3B needs about 3.0 GB of memory at its main file (a 1.8 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.8 GB to 2.6 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 3.0 GB.
Sakura Nanbeige4.2 3B – compact agentic model with strong reasoning
Sakura Nanbeige4.2 3B is a 4.17B‑parameter model built on the Nanbeige architecture. It features a LoopSplit transformer that reuses layers to increase capacity while keeping non‑embedding parameters at 3B. The model excels in agentic tasks, coding, and mathematical reasoning.
Good at
- Agentic tool‑use
- Code generation
- Mathematical reasoning
Limits
- No vision support
- Requires large context (262144 tokens)
- Apache‑2.0 license
Best for
- Personal assistant workflows
- Office automation
- Research support
Who it’s for. Developers and researchers who need a lightweight yet capable agentic model for text‑only tasks. The model is suitable for local deployment with sufficient VRAM and RAM to handle its context size.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 1.8 GB | 3.0 GB | Fits | Get file → | |
| — | 2.6 GB | 3.8 GB | Fits | Get file → |
For 8 GB VRAM, use the 1.78 GB file which estimates 3 GB RAM at 8 k context. If you have 12 GB VRAM, the 2.55 GB file is suitable, estimating 3.8 GB RAM. Both files are quantized from BF16 weights and are compatible with llama.cpp.
What GPU do you need for Sakura Nanbeige4.2 3B?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is the main file (2.6 GB file, about 3.8 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Sakura Nanbeige4.2 3B
How much VRAM does Sakura Nanbeige4.2 3B need?
About 3.0 GB at its main file with an 8K context window. That is the 1.8 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Sakura Nanbeige4.2 3B on an 8 GB GPU?
Yes, with the smaller files: the 2.6 GB file needs about 3.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Sakura Nanbeige4.2 3B GGUF?
It depends on the quant: file is 1.8 GB, file is 2.6 GB.
What context length does Sakura Nanbeige4.2 3B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 3.0 GB at 32K.
Where can I download Sakura Nanbeige4.2 3B GGUF?
The files are on Hugging Face in the webmp3/Sakura-Nanbeige4.2-3B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Sakura Nanbeige4.2 3B use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Sakura Nanbeige4.2 3B on 8 GB VRAM?
Yes, the 1.78 GB GGUF file requires about 3 GB RAM at 8 k context, which fits within 8 GB VRAM when using a 2‑stage pipeline.
Does the model support image inputs?
No, the model has no vision capability and is text‑only.
What license applies to this model?
The model is released under the Apache‑2.0 license.