Instructions to use agentlans/Llama3-ja with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use agentlans/Llama3-ja with Transformers:

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

pipe = pipeline("text-generation", model="agentlans/Llama3-ja")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("agentlans/Llama3-ja")
model = AutoModelForCausalLM.from_pretrained("agentlans/Llama3-ja")
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 agentlans/Llama3-ja with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "agentlans/Llama3-ja"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "agentlans/Llama3-ja",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/agentlans/Llama3-ja

SGLang

How to use agentlans/Llama3-ja 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 "agentlans/Llama3-ja" \
    --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": "agentlans/Llama3-ja",
		"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 "agentlans/Llama3-ja" \
        --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": "agentlans/Llama3-ja",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use agentlans/Llama3-ja with Docker Model Runner:
```
docker model run hf.co/agentlans/Llama3-ja
```

Llama3-ja

English

Model Details

This model is a linear merge of multiple Llama 3 8B models fine-tuned for Japanese language tasks, created using mergekit. The aim is to create a more robust and versatile Japanese language model that leverages the strengths of each individual model.

Intended Use

This model is designed for various Japanese natural language processing tasks, including but not limited to:

Text generation
Conversation and chatbot applications
Text completion
Question answering
Summarization

Limitations

While this model combines multiple Japanese-focused Llama 3 models, it may still have limitations:

Performance on specific tasks may vary
The model may inherit biases from its constituent models

Included models

By combining these models, we aim to create a more robust and versatile Japanese language model that leverages the strengths of each individual model.