Instructions to use shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ")
model = AutoModelForCausalLM.from_pretrained("shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ")
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 shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ

SGLang

How to use shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ 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 "shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ" \
    --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": "shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ",
		"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 "shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ" \
        --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": "shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ with Docker Model Runner:
```
docker model run hf.co/shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ
```

The Quantized LLaMA 3.1 70B Instruct Model

Original Base Model: meta-llama/Meta-Llama-3.1-70B.
Link: https://huggingface.co/meta-llama/Meta-Llama-3.1-70B

Special Notice

Please note that this is a relatively smaller model by setting group_size=1024. For the standard group_size=128 model, please check here, shuyuej/Meta-Llama-3.1-70B-GPTQ: https://huggingface.co/shuyuej/Meta-Llama-3.1-70B-GPTQ

Quantization Configurations

"quantization_config": {
    "bits": 4,
    "checkpoint_format": "gptq",
    "damp_percent": 0.01,
    "desc_act": true,
    "group_size": 1024,
    "model_file_base_name": null,
    "model_name_or_path": null,
    "quant_method": "gptq",
    "static_groups": false,
    "sym": true,
    "true_sequential": true
  },

Source Codes

Source Codes: https://github.com/vkola-lab/medpodgpt/tree/main/quantization.

Downloads last month: 7

Safetensors

Model size

71B params

Tensor type

I32

F16

Collection including shuyuej/Meta-Llama-3.1-70B-Instruct-Smaller-GPTQ

Quantization

Collection

A collection of quantized models. All the models can be fine-tuned by adding a LoRA Adapter. • 82 items • Updated Mar 2 • 3