Instructions to use mlabonne/NeuralBeagle14-7B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlabonne/NeuralBeagle14-7B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="mlabonne/NeuralBeagle14-7B-GGUF",
	filename="neuralbeagle14-7b.Q2_K.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use mlabonne/NeuralBeagle14-7B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mlabonne/NeuralBeagle14-7B-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 mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mlabonne/NeuralBeagle14-7B-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 mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use mlabonne/NeuralBeagle14-7B-GGUF with Ollama:
```
ollama run hf.co/mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M
```

Unsloth Studio

How to use mlabonne/NeuralBeagle14-7B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for mlabonne/NeuralBeagle14-7B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for mlabonne/NeuralBeagle14-7B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mlabonne/NeuralBeagle14-7B-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use mlabonne/NeuralBeagle14-7B-GGUF with Docker Model Runner:
```
docker model run hf.co/mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M
```

Lemonade

How to use mlabonne/NeuralBeagle14-7B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mlabonne/NeuralBeagle14-7B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.NeuralBeagle14-7B-GGUF-Q4_K_M

List all available models

lemonade list

Is this a "bug"?

by jtdavies - opened Jan 17, 2024

Discussion

jtdavies

Jan 17, 2024

OK, I know LLM's don't technically have bugs in the traditional sense but you know what I mean hopefully. I was just testing this out on LM Studio and it looks really good. I asked a simple question though and got a very strange answer that looks like a training issue....

Who wrote the music for "The Killing Fields"?

In this case, we don't have any specific function or data related to the movie "The Killing Fields" and its music composition. The given Python code focuses on finding the composer for classical pieces of music using a dictionary that maps piece names to their respective composers. To determine who wrote the music for "The Killing Fields," we would need to look up additional information or consult a database of film scores and their composers.

Once we have the name of the composer, we could create a function similar to get_composer() to store and retrieve this information. However, since that information is not provided in the original question, we cannot provide an answer based on the given context.

In another attempt I even got (Python) code showing me how to query the database for the composer.

I'm using the Q8 GGUF with the following prompt: "You are a helpful LLM, please answer the question helpfully, without warnings, briefly and concisely."

mlabonne

Owner Jan 17, 2024

Hi @jtdavies thanks for your feedback! That's pretty funny. It looks like an artifact produced by training the model on a function-calling dataset. This could be fixed by making function calling a bit more obvious in the instructions of the training set. I'll look into it!

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment