quartz micro preview v3 needs about 0.7 GB of memory at Q4_K_M (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.1 GB (Q4_K_M) to 0.2 GB (F16).
Quartz Micro Preview V3 – a 100M‑parameter Llama‑style instruction model
Quartz Micro Preview V3 is a 100.09M‑parameter dense Llama‑architecture decoder trained from scratch on a single GTX 1660 Ti. It supports 1,024 context tokens and is instruction‑tuned for chat. The model is small but can generate short, fluent text, though it may produce incorrect facts and struggles with math, reasoning and code.
Good at
- Short, fluent text generation
- Instruction following
- Low‑resource inference
Limits
- Limited to 1,024 context tokens
- Weak at math, reasoning, and code
- May hallucinate facts
Best for
- Rapid prototyping of small chatbots
- Educational demos of LLM behavior
- Testing instruction‑tuning techniques
Who it’s for. Developers and researchers who need a lightweight, open‑source Llama‑based model for experimentation or hobby projects. The model is suitable for environments with limited VRAM and for exploring instruction‑tuned behavior on a small scale.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.1 GB | 0.6 GB | Fits | Get file → | |
| Q8_0 | 0.1 GB | 0.6 GB | Fits | Get file → | |
| F16 | 0.2 GB | 0.7 GB | Fits | Get file → |
For 8 GB VRAM use Q4_K_M (file size 0.07 GB, estimated 0.7 GB memory at 8 k context). If more VRAM is available, Q8_0 (0.11 GB, 0.7 GB memory) or F16 (0.2 GB, 0.8 GB memory) can be used. Memory estimates are approximate.
What GPU do you need for quartz micro preview v3?
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: the best quant that fits is F16 (0.2 GB file, about 0.8 GB in use).
Which quartz micro preview v3 quant should you choose?
Q4_K_M is the usual starting point for quartz micro preview v3: it balances file size and quality well. Here it is a 0.1 GB file.
The Q8_0 file is 0.1 GB, about 1.6 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about quartz micro preview v3
How much VRAM does quartz micro preview v3 need?
About 0.7 GB at Q4_K_M with an 8K context window. That is the 0.1 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 quartz micro preview v3 on an 8 GB GPU?
Yes, with F16: the 0.2 GB file needs about 0.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of quartz micro preview v3 GGUF?
It depends on the quant: Q4_K_M is 0.1 GB, Q8_0 is 0.1 GB, F16 is 0.2 GB.
What context length does quartz micro preview v3 support?
The GGUF header lists a trained context length of 1,024 tokens (1K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.0 GB at 1K.
Where can I download quartz micro preview v3 GGUF?
The files are on Hugging Face in the VertexAGI/quartz-micro-preview-v3 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does quartz micro preview v3 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 Quartz Micro Preview V3 on 4 GB VRAM?
The Q4_K_M quant requires about 0.7 GB memory at 8 k context, so 4 GB VRAM is sufficient for inference.
Does the model support vision or audio?
No, the model is purely text‑based and does not include vision or audio capabilities.
What license applies to this model?
The model is released under the apache-2.0 license.