Nousresearch DeepHermes 3 Llama 3 3B Preview GGUF

🚀 DeepHermes-3-Llama-3-3B-Preview Quantizations

This project provides Llamacpp imatrix quantizations of the DeepHermes-3-Llama-3-3B-Preview model by NousResearch. It offers various quantization types with different file sizes and characteristics, suitable for different usage scenarios.

🚀 Quick Start

Run with LM Studio

You can run the quantized models in LM Studio.

Run with llama.cpp

You can also 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 BF16, Q8_0, Q6_K_L, etc., to meet different requirements for file size and quality.
• Online Repacking: Some quantization types support online repacking of weights, which can improve performance on ARM and AVX machines.
• Prompt Format: Defines a specific prompt format for interacting with the model.

📦 Installation

Install huggingface-cli

First, make sure you have huggingface-cli installed:

pip install -U "huggingface_hub[cli]"

Download a Specific File

Then, you can target the specific file you want:

huggingface-cli download bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF --include "NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q4_K_M.gguf" --local-dir ./

Download Split Files

If the model is bigger than 50GB and has been split into multiple files, to download them all to a local folder, run:

huggingface-cli download bartowski/NousResearch_DeepHermes-3-Llama-3-3B-Preview-GGUF --include "NousResearch_DeepHermes-3-Llama-3-3B-Preview-Q8_0/*" --local-dir ./

💻 Usage Examples

Prompt Format

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>

{prompt}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

📚 Documentation

Download a File

You can download a file (not the whole branch) from below:

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.

🔧 Technical Details

• Quantization Method: Using llama.cpp release b4877 for quantization. All quants are made using the imatrix option with a dataset from here.
• Original Model: The original model can be found at https://huggingface.co/NousResearch/DeepHermes-3-Llama-3-3B-Preview.

📄 License

This project is licensed under the llama3 license.