d3-27B Artemis 12.3GB needs about 14.9 GB of memory at Q3_K_S (a 12.3 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 12.3 GB (Q3_K_S) to 14.9 GB (IQ4_XS). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
d3-27B Artemis – a 27B decision model for text and image queries
d3-27B Artemis is a 27B multimodal decision model based on the qwen35 architecture. It answers typed questions with probabilities in a single call and supports image and video inputs. The model is licensed under apache-2.0.
Good at
- Multimodal decision making
- Fast inference on 24‑GB GPUs
- Single‑pass probability outputs
Limits
- No text generation
- Requires llama.cpp b11572 or newer
- No official support for vision beyond images/videos
Best for
- Decision‑making APIs
- Batching typed queries
- Low‑latency inference on 24‑GB GPUs
Who it’s for. Developers building decision‑oriented services that need fast, single‑pass answers for text, images, or videos. The model is suitable for applications where probability outputs are required without text generation.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K_S | 12.3 GB | 14.9 GB | Too big | Get file → | |
| IQ4_XS | 14.9 GB | 17.6 GB | Too big | Get file → |
For a 24‑GB GPU, the IQ4_XS quant (14.95 GB) fits comfortably and uses about 17.6 GB of VRAM at 8k context. The Q3_K_S quant (12.26 GB) uses about 14.9 GB of VRAM and saves additional memory if needed. Both are approximate estimates.
What GPU do you need for d3-27B Artemis 12.3GB?
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 Q3_K_S (12.3 GB file, about 14.9 GB in use).
- 24 GB VRAM: the best quant that fits is IQ4_XS (14.9 GB file, about 17.6 GB in use).
Which d3-27B Artemis 12.3GB quant should you choose?
Q3_K_S (12.3 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about d3-27B Artemis 12.3GB
How much VRAM does d3-27B Artemis 12.3GB need?
About 14.9 GB at Q3_K_S with an 8K context window. That is the 12.3 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 d3-27B Artemis 12.3GB 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 d3-27B Artemis 12.3GB GGUF?
It depends on the quant: Q3_K_S is 12.3 GB, IQ4_XS is 14.9 GB.
What context length does d3-27B Artemis 12.3GB support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download d3-27B Artemis 12.3GB GGUF?
The files are on Hugging Face in the turintech/d3-27B-Artemis-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 d3-27B Artemis 12.3GB 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 d3-27B Artemis on an 8‑GB GPU?
The smallest quant requires 12.26 GB of VRAM, so an 8‑GB GPU is insufficient.
Does the model support text generation?
No, it is a decision model that returns probabilities, not generated text.
What llama.cpp version is needed?
Requires llama.cpp release b11572 or newer, tested at commit fcac190.