gemma 4 12b constraintkit needs about 23.2 GB of memory at Q6_K (a 9.8 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.
Gemma 4 12B fine‑tuned for constraint‑kit text‑to‑CAD
Gemma 4 12B constraint‑kit is a 11.9 billion‑parameter model fine‑tuned to generate deterministic CAD programs from plain English. It writes parts, parameters, and mates that constraint‑kit compiles into real B‑rep solids. The model is reasoning‑capable and instruction‑tuned.
Good at
- CAD program generation
- Deterministic geometry output
- Reasoning tasks
Limits
- Requires ~23.2 GB VRAM at 8 k context
- File size 9.79 GB
- Text‑only input
Best for
- Generating CAD programs
- Automating design workflows
- Verifying geometry constraints
Who it’s for. Engineers, CAD designers, and developers who need an AI that writes reliable CAD code from natural language. The model is open‑source under Apache 2.0, making it suitable for research and commercial use.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q6_K | 9.8 GB | 23.2 GB | Too big | Get file → |
The only available quant is Q6_K with a 9.79 GB file. It needs about 23.2 GB VRAM at 8 k context. For lower VRAM budgets, no other quant is provided.
What GPU do you need for gemma 4 12b constraintkit?
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 Q6_K (9.8 GB file, about 23.2 GB in use).
Which gemma 4 12b constraintkit quant should you choose?
Q6_K (9.8 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about gemma 4 12b constraintkit
How much VRAM does gemma 4 12b constraintkit need?
About 23.2 GB at Q6_K with an 8K context window. That is the 9.8 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 gemma 4 12b constraintkit on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 23.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of gemma 4 12b constraintkit GGUF?
It depends on the quant: Q6_K is 9.8 GB.
What context length does gemma 4 12b constraintkit 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 gemma 4 12b constraintkit GGUF?
The files are on Hugging Face in the jajmangold/gemma-4-12b-constraintkit-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 gemma 4 12b constraintkit 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 this model on 8 GB VRAM?
No. The Q6_K quant requires about 23.2 GB VRAM at 8 k context.
Does it support image or audio input?
The fine‑tuned model is text‑only; no multimodal support is provided.
Is the model open source?
Yes, it is released under the Apache 2.0 license.