Instructions to use meta-llama/Meta-Llama-3-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meta-llama/Meta-Llama-3-8B with Transformers:

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

pipe = pipeline("text-generation", model="meta-llama/Meta-Llama-3-8B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B")
model = AutoModelForCausalLM.from_pretrained("meta-llama/Meta-Llama-3-8B", device_map="auto")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use meta-llama/Meta-Llama-3-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "meta-llama/Meta-Llama-3-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-llama/Meta-Llama-3-8B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/meta-llama/Meta-Llama-3-8B

SGLang

How to use meta-llama/Meta-Llama-3-8B 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 "meta-llama/Meta-Llama-3-8B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-llama/Meta-Llama-3-8B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "meta-llama/Meta-Llama-3-8B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-llama/Meta-Llama-3-8B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use meta-llama/Meta-Llama-3-8B with Docker Model Runner:
```
docker model run hf.co/meta-llama/Meta-Llama-3-8B
```

A noob trying to run the model with the transformers library

#125

by Alain44 - opened May 9, 2024

Discussion

Alain44

May 9, 2024

I'm a newbie with LLM.
I've tried to use the code written in the model card to launch the model on my laptop running Ubuntu (with an nvidia A500 with 4GB of RAM, CPU AMD Ryzen™ 7 7840HS 8cores/16threads, 32 GB of DDR5).
pipeline = transformers.pipeline("text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto")

When I type the code :
pipeline("what is LLM ?")
It takes forever with 1 CPU thread at 100%.

Is there a way to make this run faster or is my laptop too slow ?

Alain44

May 9, 2024

I tried with
import transformers
import torch
model_id = "meta-llama/Meta-Llama-3-8B"
pipeline = transformers.pipeline(
"text-generation",
model=model_id,
model_kwargs={
"torch_dtype": torch.bfloat16,
"quantization_config": {"load_in_4bit": True},
"low_cpu_mem_usage": True,
},
)
I've got the error : torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 28.00 MiB. GPU.

My GPU has 4GB of memory maybe it's not enough ?

BenchIt

May 10, 2024

Hey Alain,

For Llama 3 8GB, your GPU has not enough memory. Try more powerful hardware. You could start with google Colab, there you can use ressources from Google Server for free, its still not the best but should work for your 8GB Modell. Maybe also shardening could help you at this point.
If you are inferecing, choose a lower sequention length for less Ram Usage

Sry for my bad english, and GL

Ateeqq

May 12, 2024

@Alain44 try it out: https://huggingface.co/unsloth/llama-3-8b-bnb-4bit

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