Instructions to use unsloth/llama-3-8b-Instruct-bnb-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/llama-3-8b-Instruct-bnb-4bit with Transformers:

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

pipe = pipeline("text-generation", model="unsloth/llama-3-8b-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/llama-3-8b-Instruct-bnb-4bit")
model = AutoModelForCausalLM.from_pretrained("unsloth/llama-3-8b-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/llama-3-8b-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/llama-3-8b-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/llama-3-8b-Instruct-bnb-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/unsloth/llama-3-8b-Instruct-bnb-4bit

SGLang

How to use unsloth/llama-3-8b-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/llama-3-8b-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/llama-3-8b-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/llama-3-8b-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/llama-3-8b-Instruct-bnb-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use unsloth/llama-3-8b-Instruct-bnb-4bit with Docker Model Runner:
```
docker model run hf.co/unsloth/llama-3-8b-Instruct-bnb-4bit
```

BitsAndBytesConfig error

by vdavidr - opened May 1, 2024

Discussion

vdavidr

May 1, 2024

Hi, trying to load the model with

model_name = "unsloth/llama-3-8b-Instruct-bnb-4bit"
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = model_name, # Choose ANY! eg teknium/OpenHermes-2.5-Mistral-7B
max_seq_length = max_seq_length,
dtype = dtype,
load_in_4bit = load_in_4bit,
quantization_config=nf4_config,
# token = "hf_...", # use one if using gated models like meta-llama/Llama-2-7b-hf
)

I'm getting an error Unused kwargs: ['_load_in_4bit', '_load_in_8bit', 'quant_method']. These kwargs are not used in <class 'transformers.utils.quantization_config.BitsAndBytesConfig'>.

I'm using transformers-4.40.1.
I don't get the error with transformers-4.39.3.

I see the config.json uses these kwargs

danielhanchen

Unsloth AI org May 13, 2024

Oh no need to worry on the error - just ignore it - it should still work - I might update all of them to use the new kwargs

bcartel1

May 13, 2024

Hi There,
i have downloaded this model to use with ollama, i was wondering if you could leave a note on how to do this, And if not possible with ollama then how should i run it. i am new at this as you can tell :) .
thank you

shimmyshimmer

Unsloth AI org May 15, 2024

Hi There,
i have downloaded this model to use with ollama, i was wondering if you could leave a note on how to do this, And if not possible with ollama then how should i run it. i am new at this as you can tell :) .
thank you

We have a Wiki which details a bit about Ollama: https://github.com/unslothai/unsloth/wiki#use-unsloth-lora-adapter-with-ollama

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