Winnow-12B needs about 20.8 GB of memory at Q4_K_M (a 7.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 51.5 GB.
Winnow-12B Q4_K_M fine‑tune of Gemma 4 12B for typed decisions
Winnow-12B is a Gemma 4 12B IT fine‑tune for local typed decisions, chat, and image input. The GGUF download contains the merged fine‑tune; no separate adapter or base‑model download is needed.
Good at
- Typed decision scoring
- Chat completions
- Decision quality benchmarks
Limits
- No vision support in this quantization
- Requires full GPU offload for full context
- License is Apache-2.0
Best for
- Decision‑making APIs
- Low‑latency chat on laptops
- Local inference with limited VRAM
Who it’s for. Developers needing a lightweight Gemma 4 variant for decision tasks. The model is suitable for local deployment on GPUs with 12 GB VRAM or more, using full offload to keep VRAM usage low.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 7.4 GB | 20.8 GB | Too big | Get file → |
The Q4_K_M quantization is 7.38 GB and requires about 20.8 GB VRAM at 8 k context. It fits a 12 GB GPU with full offload; Ollama loads it in 8.0 GB. For larger VRAM budgets consider higher‑precision weights, but they are not listed here.
What GPU do you need for Winnow-12B?
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_K_M (7.4 GB file, about 20.8 GB in use).
Which Winnow-12B quant should you choose?
Q4_K_M is the usual starting point for Winnow-12B: it balances file size and quality well. Here it is a 7.4 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Winnow-12B
How much VRAM does Winnow-12B need?
About 20.8 GB at Q4_K_M with an 8K context window. That is the 7.4 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 Winnow-12B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 20.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Winnow-12B GGUF?
It depends on the quant: Q4_K_M is 7.4 GB.
What context length does Winnow-12B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 12.9 GB at 8K and 51.5 GB at 32K.
Where can I download Winnow-12B GGUF?
The files are on Hugging Face in the Piotr1215/Winnow-12B-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 Winnow-12B 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 Winnow-12B Q4_K_M on 8 GB VRAM?
Yes, Ollama loads the 7.38 GB model into 8.0 GB VRAM with full offload, making it compatible with 8 GB GPUs.
Does this quantization support vision?
No, vision was not tested with the Q4_K_M quantization; only text is supported.