Instructions to use cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 with Transformers:

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

pipe = pipeline("text-generation", model="cyberagent/Llama-3.1-70B-Japanese-Instruct-2407")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("cyberagent/Llama-3.1-70B-Japanese-Instruct-2407")
model = AutoModelForMultimodalLM.from_pretrained("cyberagent/Llama-3.1-70B-Japanese-Instruct-2407")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cyberagent/Llama-3.1-70B-Japanese-Instruct-2407"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cyberagent/Llama-3.1-70B-Japanese-Instruct-2407",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cyberagent/Llama-3.1-70B-Japanese-Instruct-2407

SGLang

How to use cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 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 "cyberagent/Llama-3.1-70B-Japanese-Instruct-2407" \
    --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": "cyberagent/Llama-3.1-70B-Japanese-Instruct-2407",
		"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 "cyberagent/Llama-3.1-70B-Japanese-Instruct-2407" \
        --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": "cyberagent/Llama-3.1-70B-Japanese-Instruct-2407",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cyberagent/Llama-3.1-70B-Japanese-Instruct-2407 with Docker Model Runner:
```
docker model run hf.co/cyberagent/Llama-3.1-70B-Japanese-Instruct-2407
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Llama-3.1-70B-Japanese-Instruct-2407

Model Description

This is a Japanese continually pre-trained model based on meta-llama/Meta-Llama-3.1-70B-Instruct.

Usage

Make sure to update your transformers installation via pip install --upgrade transformers.

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

model = AutoModelForCausalLM.from_pretrained("cyberagent/Llama-3.1-70B-Japanese-Instruct-2407", device_map="auto", torch_dtype="auto")
tokenizer = AutoTokenizer.from_pretrained("cyberagent/Llama-3.1-70B-Japanese-Instruct-2407")
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

messages = [
    {"role": "user", "content": "AIによって私たちの暮らしはどのように変わりますか？"}
]

input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
output_ids = model.generate(input_ids,
                            max_new_tokens=1024,
                            streamer=streamer)

Prompt Format

Llama 3.1 Format

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>

{{ user_message_1 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

{{ assistant_message_1 }}<|eot_id|><|start_header_id|>user<|end_header_id|>

{{ user_message_2 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

License

Meta Llama 3.1 Community License

Author

Ryosuke Ishigami

How to cite

@misc{cyberagent-llama-3.1-70b-japanese-instruct-2407,
      title={cyberagent/Llama-3.1-70B-Japanese-Instruct-2407},
      url={https://huggingface.co/cyberagent/Llama-3.1-70B-Japanese-Instruct-2407},
      author={Ryosuke Ishigami},
      year={2024},
}

Citations

@article{llama3.1modelcard,
    title = {Llama 3.1 Model Card},
    author = {AI@Meta},
    year = {2024},
    url = {https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/MODEL_CARD.md}
}

Downloads last month: 120

Safetensors

Model size

71B params

Tensor type

BF16

Model tree for cyberagent/Llama-3.1-70B-Japanese-Instruct-2407

Adapters

2 models

Merges

4 models

Quantizations

2 models