shingi 27b needs about 9.9 GB of memory at its main file (a 7.2 GB file) with an 8K context window, so it fits on a 12 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
Shingi 27B – ternary decision model with vision support
Shingi 27B is a 27‑billion‑parameter ternary GGUF model for local decision making. It accepts text and images, returns a choice or score without generating text. The model is a single 7.21 GB file (PQ2_0) and requires no adapters.
Good at
- Decision making with image input
- Fast inference on single GPU
- Low‑bit ternary weights
Limits
- No text generation
- Requires 9.9 GB VRAM at 8k context
- Vision tower is separate
Best for
- Rapid image‑based QA
- Embedded decision engines
- Low‑latency inference on single GPU
Who it’s for. Developers needing a lightweight, local decision model that can process images and text. Suitable for edge devices or single‑GPU deployments where memory is limited but inference speed is critical.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 7.2 GB | 9.9 GB | Too big | Get file → |
The only available quantization is PQ2_0, 7.21 GB. It requires approximately 9.9 GB VRAM at 8k context. For 8 GB VRAM, the model will not fit; consider a higher‑capacity GPU or reducing context size.
What GPU do you need for shingi 27b?
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: the best quant that fits is the main file (7.2 GB file, about 9.9 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about shingi 27b
How much VRAM does shingi 27b need?
About 9.9 GB at its main file with an 8K context window. That is the 7.2 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 shingi 27b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of shingi 27b GGUF?
It depends on the quant: file is 7.2 GB.
What context length does shingi 27b 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 shingi 27b GGUF?
The files are on Hugging Face in the kortexa-ai/shingi-27b repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does shingi 27b 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 Shingi 27B on 8 GB VRAM?
No. The model needs about 9.9 GB VRAM at 8k context, so 8 GB is insufficient.
What GPU is recommended for Shingi 27B?
An NVIDIA RTX 4090 or similar GPU with at least 10 GB VRAM provides smooth performance.
Does Shingi 27B support Apple Silicon?
Yes, it can run on Apple Silicon via the MLX companion, but the vision tower is a separate Q8_0 mmproj pack.