Katanemo Arch Function Chat 3B GGUF

🚀 Llamacpp imatrix Quantizations of Arch-Function-Chat-3B by katanemo

This project provides Llama.cpp imatrix quantizations of the Arch-Function-Chat-3B model by katanemo. It offers various quantization options for different performance and quality requirements, allowing users to run the model efficiently on different hardware.

📦 Metadata

🚀 Quick Start

• Quantization Tool: Using llama.cpp release b5010 for quantization.
• Original Model: https://huggingface.co/katanemo/Arch-Function-Chat-3B
• All Quants: All quants are made using the imatrix option with the dataset from here.
• Running Options:
  • Run them in LM Studio.
  • Run them directly with llama.cpp, or any other llama.cpp based project.

💻 Usage Examples

Basic Usage

The prompt format for this model is as follows:

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

📚 Documentation

Download a File

You can download a specific file (not the whole branch) 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.

Downloading using huggingface-cli

pip install -U "huggingface_hub[cli]"

huggingface-cli download bartowski/katanemo_Arch-Function-Chat-3B-GGUF --include "katanemo_Arch-Function-Chat-3B-Q4_K_M.gguf" --local-dir ./

huggingface-cli download bartowski/katanemo_Arch-Function-Chat-3B-GGUF --include "katanemo_Arch-Function-Chat-3B-Q8_0/*" --local-dir ./

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?

🔧 Technical Details

The project uses llama.cpp for quantization, and the specific release version is b5010. The quantization process uses the imatrix option with a calibration dataset from here.

📄 License

This project is released under the katanemo-research 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