Tesslate Tessa T1 3B GGUF

🚀 Llamacpp imatrix Quantizations of Tessa-T1-3B by Tesslate

This project offers quantized versions of the Tessa-T1-3B model using the llama.cpp library. These quantized models provide a balance between performance and resource usage, making them suitable for a variety of environments.

🚀 Quick Start

Prerequisites

• Ensure you have the necessary dependencies installed, such as huggingface-cli if you plan to download the model files.

Running the Model

• You can run the quantized models in LM Studio.
• Alternatively, you can use llama.cpp or any other llama.cpp-based project.

✨ Features

Quantization

• Using llama.cpp release b4978 for quantization.
• All quants are made using the imatrix option with a dataset from here.

Prompt Format

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

📦 Installation

Downloading using huggingface-cli

pip install -U "huggingface_hub[cli]"

huggingface-cli download bartowski/Tesslate_Tessa-T1-3B-GGUF --include "Tesslate_Tessa-T1-3B-Q4_K_M.gguf" --local-dir ./

huggingface-cli download bartowski/Tesslate_Tessa-T1-3B-GGUF --include "Tesslate_Tessa-T1-3B-Q8_0/*" --local-dir ./

💻 Usage Examples

Downloading a Specific File

huggingface-cli download bartowski/Tesslate_Tessa-T1-3B-GGUF --include "Tesslate_Tessa-T1-3B-Q4_K_M.gguf" --local-dir ./

Downloading Split Files

huggingface-cli download bartowski/Tesslate_Tessa-T1-3B-GGUF --include "Tesslate_Tessa-T1-3B-Q8_0/*" --local-dir ./

📚 Documentation

Download a file (not the whole branch)

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?

🔧 Technical Details

This project uses the llama.cpp library for quantization. The imatrix option is used to create the calibration dataset, which helps in achieving better quantization results. The online repacking feature for weights in Q4_0 models improves performance on ARM and AVX machines.

📄 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