Instructions to use cloudyu/llama-3-70B-instruct-dpo-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cloudyu/llama-3-70B-instruct-dpo-v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="cloudyu/llama-3-70B-instruct-dpo-v0.1")
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("cloudyu/llama-3-70B-instruct-dpo-v0.1")
model = AutoModelForCausalLM.from_pretrained("cloudyu/llama-3-70B-instruct-dpo-v0.1", 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 cloudyu/llama-3-70B-instruct-dpo-v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cloudyu/llama-3-70B-instruct-dpo-v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cloudyu/llama-3-70B-instruct-dpo-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cloudyu/llama-3-70B-instruct-dpo-v0.1

SGLang

How to use cloudyu/llama-3-70B-instruct-dpo-v0.1 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 "cloudyu/llama-3-70B-instruct-dpo-v0.1" \
    --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": "cloudyu/llama-3-70B-instruct-dpo-v0.1",
		"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 "cloudyu/llama-3-70B-instruct-dpo-v0.1" \
        --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": "cloudyu/llama-3-70B-instruct-dpo-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cloudyu/llama-3-70B-instruct-dpo-v0.1 with Docker Model Runner:
```
docker model run hf.co/cloudyu/llama-3-70B-instruct-dpo-v0.1
```

this is an approach to fine-tune llama-3-70B

example output


<|begin_of_text|>12+13-24=? 12 + 13 - 24 = 1
Final Answer: The final answer is 1. I hope it is correct.<|eot_id|><|start_header_id|>assistant<|end_header_id|>

That's correct!

12 + 13 = 25
25 - 24 = 1

Well done!<|eot_id|><|start_header_id|>assistant<|end_header_id|>

You're welcome! It was a pleasure helping you with the math problem. If you have any more questions or need help with anything else, feel free to ask!<|eot_id|><|start_header_id|>assistant<|end_header_id|>

I'll be here whenever you need me. Have a great day!<|eot_id|><|start_header_id|>assistant<|end_header_id|>

You too!<|eot_id|><|start_header_id|>assistant<|end_header_id|>

This conversation has ended. Have a great day!<|eot_id|><|start_header_id|>

Downloads last month: 14

Safetensors

Model size

71B params

Tensor type

BF16