LGAI EXAONE EXAONE Deep 2.4B GGUF

🚀 Llamacpp imatrix Quantizations of EXAONE-Deep-2.4B by LGAI-EXAONE

This project provides Llama.cpp imatrix quantizations of the EXAONE-Deep-2.4B model by LGAI-EXAONE. It enables efficient use of the model with different quantization types, offering a range of options for various performance and quality requirements.

🚀 Quick Start

• Quantization Method: Using llama.cpp release b4915 for quantization.
• Original Model: You can find the original model at EXAONE-Deep-2.4B.
• Running the Model:
  • You can run the quantized models in LM Studio.
  • Or run them directly with llama.cpp, or any other llama.cpp based project.

✨ Features

• Multiple Quantization Types: Offers a wide range of quantization types, including bf16, Q8_0, Q6_K_L, etc., to meet different performance and quality needs.
• Embed/Output Weights Optimization: Some quantizations (e.g., Q3_K_XL, Q4_K_L) use Q8_0 for embed and output weights, improving quality.
• Online Repacking: Supports online repacking of weights for ARM and AVX machines, enhancing performance.

📦 Installation

Download a file (not the whole branch)

You can choose the file you need from the following table:

Downloading using huggingface-cli

pip install -U "huggingface_hub[cli]"

huggingface-cli download bartowski/LGAI-EXAONE_EXAONE-Deep-2.4B-GGUF --include "LGAI-EXAONE_EXAONE-Deep-2.4B-Q4_K_M.gguf" --local-dir ./

huggingface-cli download bartowski/LGAI-EXAONE_EXAONE-Deep-2.4B-GGUF --include "LGAI-EXAONE_EXAONE-Deep-2.4B-Q8_0/*" --local-dir ./

💻 Usage Examples

Prompt format

[|system|]{system_prompt}[|endofturn|]
[|user|]{prompt}
[|assistant|]<thought>

📚 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.

Which file should I choose?

🔧 Technical Details

• Quantization Method: Uses the imatrix option in Llama.cpp for quantization, with a calibration dataset from here.
• Embed/Output Weights: Some quantizations use Q8_0 for embed and output weights to improve quality.
• Online Repacking: Supports online repacking of weights for ARM and AVX machines, enhancing performance.

📄 License

The project is under the exaone 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