Marco Mini Instruct needs about 11.9 GB of memory at Q4_K_M (a 10.5 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 11 quants, from 6.4 GB (Q2_K) to 18.4 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.
Marco Mini Instruct – 17.3B parameter sparse MoE instruction model
Marco Mini Instruct is a 17.3B parameter sparse Mixture‑of‑Experts model tuned for instruction following. It activates 0.86B parameters per token and supports multilingual instruction tasks. The model is released under Apache‑2.0.
Good at
- Multilingual instruction following
- Reasoning tasks
- Tool usage
Limits
- No vision support
- Requires 8,192 token context
- Large VRAM for higher quant
Best for
- Chatbots needing multilingual support
- Research on sparse MoE behavior
- Low‑cost instruction fine‑tuning
Who it’s for. Developers building multilingual LLM applications who need a balance between performance and memory usage. Researchers interested in sparse MoE architectures may also find it useful.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 6.4 GB | 7.8 GB | Tight | Get file → | |
| Q3_K_S | 7.5 GB | 8.9 GB | Too big | Get file → | |
| Q3_K_M | 8.3 GB | 9.7 GB | Too big | Get file → | |
| Q3_K_L | 9.0 GB | 10.4 GB | Too big | Get file → | |
| IQ4_XS | 9.3 GB | 10.8 GB | Too big | Get file → | |
| Q4_K_S | 9.9 GB | 11.3 GB | Too big | Get file → | |
| Q4_K_M | 10.5 GB | 11.9 GB | Too big | Get file → | |
| Q5_K_S | 11.9 GB | 13.4 GB | Too big | Get file → | |
| Q5_K_M | 12.3 GB | 13.7 GB | Too big | Get file → | |
| Q6_K | 14.2 GB | 15.6 GB | Too big | Get file → | |
| Q8_0 | 18.4 GB | 19.8 GB | Too big | Get file → |
For 8 GB VRAM, use Q2_K (file size 6.35 GB, estimated 7.8 GB memory). For 12 GB VRAM, Q4_K_S (9.85 GB, 11.3 GB memory) or Q3_K_M (8.29 GB, 9.7 GB memory). For 16 GB VRAM, Q5_K_S (11.91 GB, 13.4 GB memory) or Q6_K (14.18 GB, 15.6 GB memory). Estimates are approximate.
What GPU do you need for Marco Mini Instruct?
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: the best quant that fits is Q2_K (6.4 GB file, about 7.8 GB in use).
- 12 GB VRAM: the best quant that fits is Q4_K_M (10.5 GB file, about 11.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q6_K (14.2 GB file, about 15.6 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (18.4 GB file, about 19.8 GB in use).
Which Marco Mini Instruct quant should you choose?
Q4_K_M is the usual starting point for Marco Mini Instruct: it balances file size and quality well. Here it is a 10.5 GB file.
The Q8_0 file is 18.4 GB, about 1.7 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.
If memory is tight, the smallest file is Q2_K at 6.4 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Marco Mini Instruct
How much VRAM does Marco Mini Instruct need?
About 11.9 GB at Q4_K_M with an 8K context window. That is the 10.5 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 Marco Mini Instruct on an 8 GB GPU?
Yes, with Q2_K: the 6.4 GB file needs about 7.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Marco Mini Instruct GGUF?
It depends on the quant: Q2_K is 6.4 GB, Q3_K_S is 7.5 GB, Q3_K_M is 8.3 GB, Q3_K_L is 9.0 GB, IQ4_XS is 9.3 GB, Q4_K_S is 9.9 GB and 5 more.
What context length does Marco Mini Instruct support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.
Where can I download Marco Mini Instruct GGUF?
The files are on Hugging Face in the mradermacher/Marco-Mini-Instruct-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 Marco Mini Instruct 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 Marco Mini Instruct on 8 GB VRAM?
Yes, the Q2_K quant uses about 7.8 GB memory at 8,192 tokens, fitting within 8 GB VRAM.
Does this model support vision inputs?
No, Marco Mini Instruct is a text‑only model with no vision capability.
Is the model available for commercial use?
It is released under the Apache‑2.0 license, allowing commercial use with attribution.