Cognitivecomputations Qwen3 72B Embiggened GGUF

🚀 Llamacpp imatrix Quantizations of Qwen3-72B-Embiggened by cognitivecomputations

This project provides quantized versions of the Qwen3-72B-Embiggened model using llama.cpp, offering various quantization types for different performance and quality requirements.

🚀 Quick Start

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

✨ Features

• Multiple Quantization Types: Offers a wide range of quantization types, such as Q8_0, Q6_K, Q5_K_M, etc., to meet different performance and quality needs.
• Online Repacking: Supports "online repacking" for weights, which can automatically optimize performance on ARM and AVX machines.

📦 Installation

Prerequisites

First, make sure you have huggingface-cli installed:

pip install -U "huggingface_hub[cli]"

Download a Specific File

You can target the specific file you want:

huggingface-cli download bartowski/cognitivecomputations_Qwen3-72B-Embiggened-GGUF --include "cognitivecomputations_Qwen3-72B-Embiggened-Q4_K_M.gguf" --local-dir ./

Download Split Files

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/cognitivecomputations_Qwen3-72B-Embiggened-GGUF --include "cognitivecomputations_Qwen3-72B-Embiggened-Q8_0/*" --local-dir ./

You can either specify a new local-dir (e.g., cognitivecomputations_Qwen3-72B-Embiggened-Q8_0) or download them all in place (./).

💻 Usage Examples

Prompt Format

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

📚 Documentation

Model Information

Download Files

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, 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 increase.

Which File to Choose

🔧 Technical Details

Online Repacking

The "online repacking" feature for weights is introduced in this PR. If you use Q4_0 and your hardware would benefit from repacking weights, it will do it automatically on the fly.

Performance on ARM and AVX Machines

Previously, Q4_0_4_4/4_8/8_8 models had their weights interleaved in memory to improve performance on ARM and AVX machines. Now, with the "online repacking" feature, Q4_0 can achieve similar performance improvements automatically.

IQ4_NL for Better Quality

Thanks to this PR, IQ4_NL can provide slightly better quality and also repack the weights for ARM (only the 4_4 for now). The loading time may be slower but it will result in an overall speed increase.

📄 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.

If you want to support my work, visit my ko-fi page here: https://ko-fi.com/bartowski