Spectre 37B v1a GGUF Dirty needs about 61.3 GB of memory at Q4_K_M (a 22.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 6 quants, from 14.1 GB (Q2_K) to 38.9 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 154.6 GB.
Spectre 37B v1a GGUF Dirty – a 37 B Gemma4 model in GGUF format
Spectre 37B v1a GGUF Dirty is a 37‑billion‑parameter Gemma4 model released by BeaverAI. It is distributed in GGUF format and supports multiple quantizations. The model was trained on a context of 262,144 tokens.
Good at
- Large context handling
- High‑parameter density
- Versatile quantization options
Limits
- No vision support
- License is gemma
- Large VRAM requirement for full precision
Best for
- Long‑form text generation
- Complex reasoning tasks
- Fine‑tuning with limited VRAM
Who it’s for. Developers and researchers needing a large Gemma4 model for text‑only workloads. The model is suitable for those who can manage the memory demands and are comfortable with the gemma license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 14.1 GB | 53.2 GB | Too big | Get file → | |
| Q3_K_M | 18.1 GB | 57.3 GB | Too big | Get file → | |
| Q4_K_M | 22.2 GB | 61.3 GB | Too big | Get file → | |
| Q5_K_M | 26.0 GB | 65.1 GB | Too big | Get file → | |
| Q6_K | 30.0 GB | 69.2 GB | Too big | Get file → | |
| Q8_0 | 38.9 GB | 78.0 GB | Too big | Get file → |
For 8 GB VRAM, use Q2_K (14.07 GB file, ~53 GB memory at 8 k context). For 12 GB VRAM, Q3_K_M (18.11 GB file, ~57 GB memory). For 16 GB VRAM, Q4_K_M (22.19 GB file, ~61 GB memory).
What GPU do you need for Spectre 37B v1a GGUF Dirty?
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: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Spectre 37B v1a GGUF Dirty quant should you choose?
Q4_K_M is the usual starting point for Spectre 37B v1a GGUF Dirty: it balances file size and quality well. Here it is a 22.2 GB file.
The Q8_0 file is 38.9 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 14.1 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 Spectre 37B v1a GGUF Dirty
How much VRAM does Spectre 37B v1a GGUF Dirty need?
About 61.3 GB at Q4_K_M with an 8K context window. That is the 22.2 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 Spectre 37B v1a GGUF Dirty on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 53.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Spectre 37B v1a GGUF Dirty GGUF?
It depends on the quant: Q2_K is 14.1 GB, Q3_K_M is 18.1 GB, Q4_K_M is 22.2 GB, Q5_K_M is 26.0 GB, Q6_K is 30.0 GB, Q8_0 is 38.9 GB.
What context length does Spectre 37B v1a GGUF Dirty support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 38.7 GB at 8K and 154.6 GB at 32K.
Where can I download Spectre 37B v1a GGUF Dirty GGUF?
The files are on Hugging Face in the BeaverAI/Spectre-37B-v1a-GGUF-Dirty repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Spectre 37B v1a GGUF Dirty use?
Hugging Face lists the license as gemma. 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 Spectre 37B v1a GGUF Dirty on 8 GB VRAM?
Use the Q2_K quantization; it requires about 53 GB of VRAM at 8 k context, so 8 GB is insufficient for inference.
Does this model support image inputs?
No, the model has no vision capabilities.
What license applies to this model?
The model is licensed under the gemma license.