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

Libraries

How to use mlabonne/NeuralBeagle14-7B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="mlabonne/NeuralBeagle14-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mlabonne/NeuralBeagle14-7B")
model = AutoModelForCausalLM.from_pretrained("mlabonne/NeuralBeagle14-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mlabonne/NeuralBeagle14-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlabonne/NeuralBeagle14-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlabonne/NeuralBeagle14-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

SGLang

How to use mlabonne/NeuralBeagle14-7B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "mlabonne/NeuralBeagle14-7B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlabonne/NeuralBeagle14-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "mlabonne/NeuralBeagle14-7B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlabonne/NeuralBeagle14-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

Quantized version please

by HR1777 - opened Jan 16, 2024

Discussion

HR1777

Jan 16, 2024

I am pretty sure many users would love to download the GGUF version of @mlabonne models, but unfortunately no one shows any interests in releasing them :(

mlabonne

Owner Jan 16, 2024

Thanks! Currently making one, it's gonna be uploaded in a few minutes + a chat space to give it a try.

HR1777

Jan 16, 2024

That's great. I'm waiting for both files and chat space.

mlabonne

Owner Jan 16, 2024

Sorry it failed. It'll be uploaded in 5-6 hours

sudhir2016

Jan 16, 2024

Meanwhile you can try my humble attempt sudhir2016/NeuralBeagle14-7B-GGUF till the original work from the master himself is available !!

mlabonne

Owner Jan 16, 2024

Thanks @sudhir2016 ! The space is now available here: https://huggingface.co/spaces/mlabonne/NeuralBeagle14-7B-GGUF-Chat (GGUF: https://huggingface.co/mlabonne/NeuralBeagle14-7B-GGUF).

CultriX

Jan 17, 2024

Thanks for your efforts @mlabonne . Great model!

testerav

Jan 17, 2024

•

edited Jan 17, 2024

@mlabonne Q4_K_M or Q5_K_M for 7b models? is their amy significant difference? i see that earlier the space was running the Q5 model but you switched to Q4

HR1777

Jan 17, 2024

Thank you @mlabonne for the GGUF version. I really like it! My request is done, but I just don't close the discussion because of the questions asked by pother users.

mlabonne

Owner Jan 17, 2024

@mlabonne Q4_K_M or Q5_K_M for 7b models? is their amy significant difference? i see that earlier the space was running the Q5 model but you switched to Q4

Q5_K_M is slightly better, I changed it because the inference was too slow on a CPU.

Thank you @mlabonne for the GGUF version. I really like it! My request is done, but I just don't close the discussion because of the questions asked by pother users.

Thanks @HR1777 sure let's keep it open

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