Instructions to use Amar-89/Llama-3.1-8B-Instruct-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Amar-89/Llama-3.1-8B-Instruct-8bit with Transformers:

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

pipe = pipeline("text-generation", model="Amar-89/Llama-3.1-8B-Instruct-8bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Amar-89/Llama-3.1-8B-Instruct-8bit")
model = AutoModelForCausalLM.from_pretrained("Amar-89/Llama-3.1-8B-Instruct-8bit", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Amar-89/Llama-3.1-8B-Instruct-8bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Amar-89/Llama-3.1-8B-Instruct-8bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Amar-89/Llama-3.1-8B-Instruct-8bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Amar-89/Llama-3.1-8B-Instruct-8bit

SGLang

How to use Amar-89/Llama-3.1-8B-Instruct-8bit 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 "Amar-89/Llama-3.1-8B-Instruct-8bit" \
    --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": "Amar-89/Llama-3.1-8B-Instruct-8bit",
		"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 "Amar-89/Llama-3.1-8B-Instruct-8bit" \
        --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": "Amar-89/Llama-3.1-8B-Instruct-8bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Amar-89/Llama-3.1-8B-Instruct-8bit with Docker Model Runner:
```
docker model run hf.co/Amar-89/Llama-3.1-8B-Instruct-8bit
```

Model Card

Model Details

Developed by: Amar-89
Model type: Quantized (8-bit)
License: MIT
Quantized from model: meta-llama/Llama-3.1-8B-Instruct
Model size: 9.1 GB

Uses the tokenizer from the base model. No additional tweaks to model besides quantization. Recommended: 12 GB VRAM

How to use

pip install -q -U torch bitsandbytes transformers accelerate

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "Amar-89/Llama-3.1-8B-Instruct-8bit"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

def terminal_chat(model, tokenizer, system_prompt):
    """
    Starts a terminal-based chat session with a specified model, tokenizer, and system prompt.

    Args:
        model: The Hugging Face model object.
        tokenizer: The Hugging Face tokenizer object.
        system_prompt: The system role or instruction to define the chat behavior.
    """
    from transformers import pipeline

    pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

    messages = [{"role": "system", "content": system_prompt}]
    print("Chat session started. Type 'exit' to quit.")

    while True:
        user_input = input("User: ")
        if user_input.lower() == "exit":
            print("Ending chat session. Goodbye!")
            break

        messages.append({"role": "user", "content": user_input})

        outputs = pipe(messages, max_new_tokens=256)

        response = outputs[0]["generated_text"][-1]['content']
        print(f"Assistant: {response}")

        print(messages)


system_prompt = "You are a pirate chatbot who always responds in pirate speak!"

terminal_chat(model, tokenizer, system_prompt)

Downloads last month: 15

Safetensors

Model size

8B params

Tensor type

F32

BF16

Model tree for Amar-89/Llama-3.1-8B-Instruct-8bit

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Quantized

(922)

this model