Arcee-ai_Virtuoso-Small-v2-GGUF Open-source Model - Diverse Quantization to Meet Different Hardware Performance Requirements

Arcee Ai Virtuoso Small V2 GGUF

Developed by bartowski

A quantized version of the arcee-ai/Virtuoso-Small-v2 model based on llama.cpp, offering multiple quantization types to meet different hardware and performance requirements.

Large Language Model Open Source License:Apache-2.0 #Multiple quantization options #ARM/AVX optimization #Low-memory adaptation

Downloads 1,976

Release Time : 2/1/2025

Model Overview

This model is a quantized version of Virtuoso-Small-v2, processed by the llama.cpp tool to generate quantized model files of various types and sizes, suitable for different hardware environments and performance requirements.

Model Features

Multiple quantization types

Provide a rich variety of quantization types, such as f32, f16, Q8_0, Q6_K_L, etc., to meet different quality and performance requirements.

Online repackaging

Some quantized models support online repackaging and can automatically optimize performance on ARM and AVX machines.

Clear prompt format

The prompt format of the model is clearly defined for easy use.

Model Capabilities

Text generation

Multilingual support

Quantized model optimization

Use Cases

Text generation

Chatbot

Can be used to build chatbots, supporting multi-round conversations.

Content creation

Can be used to generate text content such as articles and stories.

🚀 Llamacpp imatrix Quantizations of Virtuoso-Small-v2 by arcee-ai

This project provides Llama.cpp imatrix quantizations of the Virtuoso-Small-v2 model by arcee-ai. It uses specific releases of Llama.cpp for quantization, and the quantized models can be run in various environments.

🚀 Quick Start

Using LM Studio: You can run the quantized models in LM Studio.
Using Llama.cpp: Run them directly with llama.cpp, or any other llama.cpp based project.

✨ Features

Quantization: Using llama.cpp release b4585 for quantization.
Original Model: The original model can be found at https://huggingface.co/arcee-ai/Virtuoso-Small-v2.
Quantization Option: All quants are made using the imatrix option with a dataset from here.

💻 Usage Examples

Prompt Format

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

📦 Installation

Download a file (not the whole branch)

Filename	Quant type	File Size	Split	Description
Virtuoso-Small-v2-f32.gguf	f32	59.07GB	true	Full F32 weights.
Virtuoso-Small-v2-f16.gguf	f16	29.54GB	false	Full F16 weights.
Virtuoso-Small-v2-Q8_0.gguf	Q8_0	15.70GB	false	Extremely high quality, generally unneeded but max available quant.
...	...	...	...	...

Downloading using huggingface-cli

Click to view download instructions

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download bartowski/arcee-ai_Virtuoso-Small-v2-GGUF --include "arcee-ai_Virtuoso-Small-v2-Q4_K_M.gguf" --local-dir ./

If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download bartowski/arcee-ai_Virtuoso-Small-v2-GGUF --include "arcee-ai_Virtuoso-Small-v2-Q8_0/*" --local-dir ./

You can either specify a new local-dir (arcee-ai_Virtuoso-Small-v2-Q8_0) or download them all in place (./)

📚 Documentation

Embed/output weights

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

ARM/AVX information

Previously, you would download Q4_0_4_4/4_8/8_8, and these would have their weights interleaved in memory in order to improve performance on ARM and AVX machines by loading up more data in one pass.

Now, however, there is something called "online repacking" for weights. details in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

As of llama.cpp build b4282 you will not be able to run the Q4_0_X_X files and will instead need to use Q4_0.

Additionally, if you want to get slightly better quality for , you can use IQ4_NL thanks to this PR which will also repack the weights for ARM, though only the 4_4 for now. The loading time may be slower but it will result in an overall speed incrase.

Click to view Q4_0_X_X information (deprecated

I'm keeping this section to show the potential theoretical uplift in performance from using the Q4_0 with online repacking.

Click to view benchmarks on an AVX2 system (EPYC7702)

model	size	params	backend	threads	test	t/s	% (vs Q4_0)
qwen2 3B Q4_0	1.70 GiB	3.09 B	CPU	64	pp512	204.03 ± 1.03	100%
...	...	...	...	...	...	...	...

Q4_0_8_8 offers a nice bump to prompt processing and a small bump to text generation

Which file should I choose?

Click here for details

A great write up with charts showing various performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

The I-quants are not compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.

📄 License

This project is licensed under the apache-2.0 license.

👏 Credits

Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset. Thank you ZeroWw for the inspiration to experiment with embed/output. Thank you to LM Studio for sponsoring my work.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご