Viking-33B needs about 22.4 GB of memory at Q4_K_M (a 20.0 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 14 quants, from 12.6 GB (Q2_K) to 35.2 GB (Q8_0).
Viking‑33B: 33B‑parameter LLaMA‑style decoder for code and Scandinavian languages
Viking‑33B is a 33.1B parameter decoder‑only transformer. It was trained on 2 trillion tokens, including Finnish, English, Swedish, Danish, Norwegian, Icelandic and code. The model is fully open source under Apache‑2.0.
Good at
- Code generation
- Scandinavian language understanding
- Large context handling
Limits
- No vision support
- Requires 8k context for estimates
- No mixture of experts
Best for
- Developers needing code assistance
- Researchers working with Scandinavian corpora
- Local inference on high‑memory GPUs
Who it’s for. The model is suitable for users who need a large, open‑source LLM capable of coding and Scandinavian language tasks. It is ideal for developers and researchers who can run it locally on GPUs with sufficient VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 12.6 GB | 14.0 GB | Too big | Get file → | |
| IQ3_XS | 13.9 GB | 15.4 GB | Too big | Get file → | |
| Q3_K_S | 14.6 GB | 16.0 GB | Too big | Get file → | |
| IQ3_S | 14.7 GB | 16.1 GB | Too big | Get file → | |
| IQ3_M | 15.2 GB | 16.6 GB | Too big | Get file → | |
| Q3_K_M | 16.2 GB | 17.6 GB | Too big | Get file → | |
| Q3_K_L | 17.6 GB | 19.0 GB | Too big | Get file → | |
| IQ4_XS | 18.1 GB | 19.5 GB | Too big | Get file → | |
| Q4_K_S | 19.0 GB | 20.5 GB | Too big | Get file → | |
| Q4_K_M | 20.0 GB | 21.4 GB | Too big | Get file → | |
| Q5_K_S | 22.9 GB | 24.3 GB | Too big | Get file → | |
| Q5_K_M | 23.5 GB | 24.9 GB | Too big | Get file → | |
| Q6_K | 27.2 GB | 28.6 GB | Too big | Get file → | |
| Q8_0 | 35.2 GB | 36.6 GB | Too big | Get file → |
For 8‑GB VRAM use Q2_K (file size 12.56 GB, ~14.9 GB memory). For 16‑GB VRAM use Q4_K_M (19.99 GB, ~22.4 GB memory). For 32‑GB VRAM use Q5_K_M (23.47 GB, ~25.9 GB memory). Estimates are approximate.
What GPU do you need for Viking-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: the best quant that fits is Q2_K (12.6 GB file, about 14.9 GB in use).
- 24 GB VRAM: the best quant that fits is Q4_K_M (20.0 GB file, about 22.4 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (27.2 GB file, about 29.6 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (35.2 GB file, about 37.6 GB in use).
Which Viking-33B quant should you choose?
Q4_K_M is the usual starting point for Viking-33B: it balances file size and quality well. Here it is a 20.0 GB file.
The Q8_0 file is 35.2 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 12.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Viking-33B
How much VRAM does Viking-33B need?
About 22.4 GB at Q4_K_M with an 8K context window. That is the 20.0 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 Viking-33B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 14.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Viking-33B GGUF?
It depends on the quant: Q2_K is 12.6 GB, IQ3_XS is 13.9 GB, Q3_K_S is 14.6 GB, IQ3_S is 14.7 GB, IQ3_M is 15.2 GB, Q3_K_M is 16.2 GB and 8 more.
What context length does Viking-33B support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 1.9 GB at 8K and 0.9 GB at 4K.
Where can I download Viking-33B GGUF?
The files are on Hugging Face in the mradermacher/Viking-33B-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 Viking-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.
Can I run Viking‑33B on 8 GB VRAM?
Use the Q2_K quant; it requires about 14.9 GB of VRAM at 8k context, which is close to 8 GB but may need off‑load or batching.
Does Viking‑33B support image input?
No, the model has no vision capabilities.
What license applies to Viking‑33B?
The model is released under the Apache‑2.0 license.