Nanbeige4.3 33B needs about 20.8 GB of memory at Q4_0 (a 18.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 4 quants, from 17.6 GB (Q4_0) to 34.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 7.0 GB.
Nanbeige4.3 33B is a 33.1B-parameter Nanbeige model in GGUF format, quantized by Harsha001. Its trained context length is 512K tokens. The repository has 4 quant files, from 17.6 GB (Q4_0) to 34.2 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 17.6 GB | 19.8 GB | Too big | Get file → | |
| Q4_0 | 18.6 GB | 20.8 GB | Too big | Get file → | |
| Q6_K | 27.3 GB | 29.6 GB | Too big | Get file → | |
| Q8_0 | 34.2 GB | 36.4 GB | Too big | Get file → |
What GPU do you need for Nanbeige4.3 33B?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is Q4_0 (18.6 GB file, about 20.8 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (27.3 GB file, about 29.6 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (34.2 GB file, about 36.4 GB in use).
Which Nanbeige4.3 33B quant should you choose?
Q4_0 (18.6 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
If memory is tight, the smallest file is Q4_0 at 17.6 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Nanbeige4.3 33B
How much VRAM does Nanbeige4.3 33B need?
About 20.8 GB at Q4_0 with an 8K context window. That is the 18.6 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 Nanbeige4.3 33B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 19.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Nanbeige4.3 33B GGUF?
It depends on the quant: Q4_0 is 17.6 GB, Q4_0 is 18.6 GB, Q6_K is 27.3 GB, Q8_0 is 34.2 GB.
What context length does Nanbeige4.3 33B support?
The GGUF header lists a trained context length of 524,288 tokens (512K). The KV cache grows with the context you use: about 1.7 GB at 8K and 7.0 GB at 32K.
Where can I download Nanbeige4.3 33B GGUF?
The files are on Hugging Face in the Harsha001/Nanbeige4.3-33B-ROCmFPX repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Nanbeige4.3 33B 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.