Instructions to use unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit with Transformers:

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

pipe = pipeline("text-generation", model="unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit")
model = AutoModelForCausalLM.from_pretrained("unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit", device_map="auto")
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 Settings

vLLM

How to use unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit

SGLang

How to use unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit 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 "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit" \
    --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": "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit",
		"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 "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit" \
        --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": "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit 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 unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit 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 unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit",
    max_seq_length=2048,
)

Docker Model Runner
How to use unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit with Docker Model Runner:
```
docker model run hf.co/unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit
```

RuntimeError: Unsloth: `unsloth/Meta-Llama-3.1-8B-bnb-4bit` is not a base model or a PEFT model.

by yorickdejong - opened Jul 24, 2024

Discussion

yorickdejong

Jul 24, 2024

Hi there, when trying to fine tune the model, I get the following error:

RuntimeError: Unsloth: unsloth/Meta-Llama-3.1-8B-bnb-4bit is not a base model or a PEFT model.
We could not locate a config.json or adapter_config.json file.
Are you certain the model name is correct? Does it actually exist?

Previous runs with Llama3.0 worked well, however, since trying llama3.1 I am receiving the error. According to the files shared in this repo, it seems that the model does contain a config.json.

Thanks!

meabo

Jul 24, 2024

Same error here

VALilly

Jul 24, 2024

same error for me too

ewre324

Jul 24, 2024

+1
Encountering the same error for the 405B model as well.

danielhanchen

Unsloth AI org Jul 24, 2024

@ewre324 @meabo @VALilly @yorickdejong Apologies on the error!
Could you all try reinstalling Unsloth?

pip uninstall unsloth -y
pip install --upgrade --force-reinstall --no-cache-dir git+https://github.com/unslothai/unsloth.git

ridwanlekan

Jul 26, 2024

@danielhanchen Thanks for sharing an approach to solving the issue by upgrading the unsloth package. Unfortunately, the issue persists after uninstalling and upgrading with the path you shared above.

Currently, I am getting this error message.

/home/devops/miniconda3/envs/emsd_demo/lib/python3.11/site-packages/peft/utils/other.py:581: UserWarning: Unable to fetch remote file due to the following error (ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response')), '(Request ID: c9ad27cd-574c-43b8-8c2f-a49f60e4050a)') - silently ignoring the lookup for the file config.json in unsloth/llama-3-8b-bnb-4bit.
warnings.warn(
/home/devops/miniconda3/envs/emsd_demo/lib/python3.11/site-packages/peft/utils/save_and_load.py:154: UserWarning: Could not find a config file in unsloth/llama-3-8b-bnb-4bit - will assume that the vocabulary was not modified.
warnings.warn(
{'train_runtime': 610.2927, 'train_samples_per_second': 5.899, 'train_steps_per_second': 0.492, 'train_loss': 0.5799148793518544, 'epoch': 10.71}

As if this isn't enough, it hasn't been possible to revert to the old package as I keep getting same error message after several trials of reinstalling unsloth

XerWandeRer

Jul 28, 2024

update Transformer.

pip install --upgrade transformers

https://github.com/unslothai/unsloth/issues/801

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