Instructions to use bartowski/Meta-Llama-3-120B-Instruct-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use bartowski/Meta-Llama-3-120B-Instruct-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

Use Docker

docker model run hf.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use bartowski/Meta-Llama-3-120B-Instruct-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bartowski/Meta-Llama-3-120B-Instruct-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bartowski/Meta-Llama-3-120B-Instruct-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

Ollama
How to use bartowski/Meta-Llama-3-120B-Instruct-GGUF with Ollama:
```
ollama run hf.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use bartowski/Meta-Llama-3-120B-Instruct-GGUF with Docker Model Runner:
```
docker model run hf.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M
```

Lemonade

How to use bartowski/Meta-Llama-3-120B-Instruct-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull bartowski/Meta-Llama-3-120B-Instruct-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Meta-Llama-3-120B-Instruct-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

bartowski commited on May 9, 2024

Commit

e8ff0d6

verified ·

1 Parent(s): 78d2668

Update README.md

Browse files

Files changed (1) hide show

README.md +18 -18

README.md CHANGED Viewed

@@ -44,24 +44,24 @@ All quants made using imatrix option with dataset provided by Kalomaze [here](ht
 | [Meta-Llama-3-120B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q6_K.gguf) | Q6_K | 100.00GB | Very high quality, near perfect, *recommended*. |
 | [Meta-Llama-3-120B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q5_K_M.gguf) | Q5_K_M | 86.21GB | High quality, *recommended*. |
 | [Meta-Llama-3-120B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q5_K_S.gguf) | Q5_K_S | 83.95GB | High quality, *recommended*. |
-| [Meta-Llama-3-120B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-Q4_K_M.gguf) | Q4_K_M |  | Good quality, uses about 4.83 bits per weight, *recommended*. |
-| [Meta-Llama-3-120B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-Q4_K_S.gguf) | Q4_K_S |  | Slightly lower quality with more space savings, *recommended*. |
-| [Meta-Llama-3-120B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ4_NL.gguf) | IQ4_NL |  | Decent quality, slightly smaller than Q4_K_S with similar performance *recommended*. |
-| [Meta-Llama-3-120B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ4_XS.gguf) | IQ4_XS |  | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
-| [Meta-Llama-3-120B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-Q3_K_L.gguf) | Q3_K_L |  | Lower quality but usable, good for low RAM availability. |
-| [Meta-Llama-3-120B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-Q3_K_M.gguf) | Q3_K_M |  | Even lower quality. |
-| [Meta-Llama-3-120B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ3_M.gguf) | IQ3_M |  | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
-| [Meta-Llama-3-120B-Instruct-IQ3_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ3_S.gguf) | IQ3_S |  | Lower quality, new method with decent performance, recommended over Q3_K_S quant, same size with better performance. |
-| [Meta-Llama-3-120B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-Q3_K_S.gguf) | Q3_K_S |  | Low quality, not recommended. |
-| [Meta-Llama-3-120B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ3_XS.gguf) | IQ3_XS |  | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
-| [Meta-Llama-3-120B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ3_XXS.gguf) | IQ3_XXS |  | Lower quality, new method with decent performance, comparable to Q3 quants. |
-| [Meta-Llama-3-120B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-Q2_K.gguf) | Q2_K |  | Very low quality but surprisingly usable. |
-| [Meta-Llama-3-120B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ2_M.gguf) | IQ2_M |  | Very low quality, uses SOTA techniques to also be surprisingly usable. |
-| [Meta-Llama-3-120B-Instruct-IQ2_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ2_S.gguf) | IQ2_S |  | Very low quality, uses SOTA techniques to be usable. |
-| [Meta-Llama-3-120B-Instruct-IQ2_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ2_XS.gguf) | IQ2_XS |  | Very low quality, uses SOTA techniques to be usable. |
-| [Meta-Llama-3-120B-Instruct-IQ2_XXS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ2_XXS.gguf) | IQ2_XXS |  | Lower quality, uses SOTA techniques to be usable. |
-| [Meta-Llama-3-120B-Instruct-IQ1_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ1_M.gguf) | IQ1_M |  | Extremely low quality, *not* recommended. |
-| [Meta-Llama-3-120B-Instruct-IQ1_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF//main/Meta-Llama-3-120B-Instruct-IQ1_S.gguf) | IQ1_S |  | Extremely low quality, *not* recommended. |
 ## Downloading using huggingface-cli

 | [Meta-Llama-3-120B-Instruct-Q6_K.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q6_K.gguf) | Q6_K | 100.00GB | Very high quality, near perfect, *recommended*. |
 | [Meta-Llama-3-120B-Instruct-Q5_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q5_K_M.gguf) | Q5_K_M | 86.21GB | High quality, *recommended*. |
 | [Meta-Llama-3-120B-Instruct-Q5_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q5_K_S.gguf) | Q5_K_S | 83.95GB | High quality, *recommended*. |
+| [Meta-Llama-3-120B-Instruct-Q4_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q4_K_M.gguf) | Q4_K_M | 73.24GB | Good quality, uses about 4.83 bits per weight, *recommended*. |
+| [Meta-Llama-3-120B-Instruct-Q4_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q4_K_S.gguf) | Q4_K_S | 69.35GB | Slightly lower quality with more space savings, *recommended*. |
+| [Meta-Llama-3-120B-Instruct-IQ4_NL.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-IQ4_NL.gguf) | IQ4_NL | 68.99GB | Decent quality, slightly smaller than Q4_K_S with similar performance *recommended*. |
+| [Meta-Llama-3-120B-Instruct-IQ4_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-IQ4_XS.gguf) | IQ4_XS | 65.25GB | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
+| [Meta-Llama-3-120B-Instruct-Q3_K_L.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q3_K_L.gguf) | Q3_K_L | 64.00GB | Lower quality but usable, good for low RAM availability. |
+| [Meta-Llama-3-120B-Instruct-Q3_K_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q3_K_M.gguf) | Q3_K_M | 58.81GB | Even lower quality. |
+| [Meta-Llama-3-120B-Instruct-IQ3_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-IQ3_M.gguf) | IQ3_M | 54.73GB | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
+| [Meta-Llama-3-120B-Instruct-IQ3_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-IQ3_S.gguf) | IQ3_S | 52.95GB | Lower quality, new method with decent performance, recommended over Q3_K_S quant, same size with better performance. |
+| [Meta-Llama-3-120B-Instruct-Q3_K_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-Q3_K_S.gguf) | Q3_K_S | 52.80GB | Low quality, not recommended. |
+| [Meta-Llama-3-120B-Instruct-IQ3_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/tree/main/Meta-Llama-3-120B-Instruct-IQ3_XS.gguf) | IQ3_XS | 50.15GB | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
+| [Meta-Llama-3-120B-Instruct-IQ3_XXS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ3_XXS.gguf) | IQ3_XXS | 47.03GB | Lower quality, new method with decent performance, comparable to Q3 quants. |
+| [Meta-Llama-3-120B-Instruct-Q2_K.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-Q2_K.gguf) | Q2_K | 45.09GB | Very low quality but surprisingly usable. |
+| [Meta-Llama-3-120B-Instruct-IQ2_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ2_M.gguf) | IQ2_M | 41.30GB | Very low quality, uses SOTA techniques to also be surprisingly usable. |
+| [Meta-Llama-3-120B-Instruct-IQ2_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ2_S.gguf) | IQ2_S | 38.02GB | Very low quality, uses SOTA techniques to be usable. |
+| [Meta-Llama-3-120B-Instruct-IQ2_XS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ2_XS.gguf) | IQ2_XS | 36.18GB | Very low quality, uses SOTA techniques to be usable. |
+| [Meta-Llama-3-120B-Instruct-IQ2_XXS.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ2_XXS.gguf) | IQ2_XXS | 32.60GB | Lower quality, uses SOTA techniques to be usable. |
+| [Meta-Llama-3-120B-Instruct-IQ1_M.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ1_M.gguf) | IQ1_M | 28.49GB | Extremely low quality, *not* recommended. |
+| [Meta-Llama-3-120B-Instruct-IQ1_S.gguf](https://huggingface.co/bartowski/Meta-Llama-3-120B-Instruct-GGUF/blob/main/Meta-Llama-3-120B-Instruct-IQ1_S.gguf) | IQ1_S | 26.02GB | Extremely low quality, *not* recommended. |
 ## Downloading using huggingface-cli