Gryphe Codex 24B Small 3.2 GGUF

🚀 Llamacpp imatrix Quantizations of Codex-24B-Small-3.2 by Gryphe

This project provides Llama.cpp imatrix quantizations of the Codex-24B-Small-3.2 model by Gryphe. It offers various quantized versions of the model, which can be used for text generation tasks. The quantizations are created using Llama.cpp and can be run in LM Studio or any Llama.cpp-based project.

✨ Features

• Multiple Quantization Types: Offers a wide range of quantization types, such as bf16, Q8_0, Q6_K_L, etc., to suit different hardware and performance requirements.
• Online Repacking: Some quantization types support online repacking of weights, which can improve performance on ARM and AVX machines.
• Easy Download: Provides instructions on how to download the quantized files using the Hugging Face CLI.

📦 Installation

The installation process mainly involves downloading the quantized model files. You can use the Hugging Face CLI to download the specific file you need.

First, make sure you have the Hugging Face CLI installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download bartowski/Gryphe_Codex-24B-Small-3.2-GGUF --include "Gryphe_Codex-24B-Small-3.2-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/Gryphe_Codex-24B-Small-3.2-GGUF --include "Gryphe_Codex-24B-Small-3.2-Q8_0/*" --local-dir ./

💻 Usage Examples

Running in LM Studio

You can run the quantized models in LM Studio. Simply load the downloaded model file in LM Studio and start using it for text generation.

Running with Llama.cpp

You can also run the models directly with Llama.cpp, or any other Llama.cpp-based project.

📚 Documentation

Prompt Format

No chat template is specified, so the default is used. This may be incorrect, so check the original model card for details.

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

Download a File

You can download a specific file from the following table:

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.

Which File Should I Choose?

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, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

🔧 Technical Details

This section provides some technical details about the quantization process and the model.

• Quantization Method: The quantizations are created using Llama.cpp release b5697.
• Calibration Dataset: All quants are made using the imatrix option with a dataset from here.

📄 License

This project is licensed under the Apache License 2.0.

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