Instructions to use cpm-ai/Llama3-Ocelot-8B-instruct-v01 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cpm-ai/Llama3-Ocelot-8B-instruct-v01 with Transformers:

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

pipe = pipeline("text-generation", model="cpm-ai/Llama3-Ocelot-8B-instruct-v01")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cpm-ai/Llama3-Ocelot-8B-instruct-v01")
model = AutoModelForCausalLM.from_pretrained("cpm-ai/Llama3-Ocelot-8B-instruct-v01", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cpm-ai/Llama3-Ocelot-8B-instruct-v01 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cpm-ai/Llama3-Ocelot-8B-instruct-v01"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cpm-ai/Llama3-Ocelot-8B-instruct-v01",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cpm-ai/Llama3-Ocelot-8B-instruct-v01

SGLang

How to use cpm-ai/Llama3-Ocelot-8B-instruct-v01 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 "cpm-ai/Llama3-Ocelot-8B-instruct-v01" \
    --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": "cpm-ai/Llama3-Ocelot-8B-instruct-v01",
		"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 "cpm-ai/Llama3-Ocelot-8B-instruct-v01" \
        --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": "cpm-ai/Llama3-Ocelot-8B-instruct-v01",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cpm-ai/Llama3-Ocelot-8B-instruct-v01 with Docker Model Runner:
```
docker model run hf.co/cpm-ai/Llama3-Ocelot-8B-instruct-v01
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Kor-LLAMA3 Model

Update @ 2024.06.05: First release of Llama3-Ocelot-8B-instruct-v01

This model card corresponds to the 8B Instruct version of the Llama-Ko model.

The train wad done on A100-80GB

Resources and Technical Documentation:

llama Model

Citation

Model Developers: frcp, nebchi, pepperonipizza97

Model Information

It is an LLM model capable of generating Korean text, trained on a pre-trained base model with high-quality Korean SFT dataset and DPO dataset.

Inputs and outputs

Input: Text string, such as a question, a prompt, or a document to be summarized.
Output: Generated Korean-language text in response to the input, such as an answer to a question, or a summary of a document.

Running the model on a single / multi GPU

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cpm-ai/Ocelot-Ko-self-instruction-10.8B-v1.0")
model = AutoModelForCausalLM.from_pretrained("cpm-ai/Ocelot-Ko-self-instruction-10.8B-v1.0", device_map="auto")

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=4096, streamer=streamer)

text = '대한민국의 수도는 어디인가요?'

messages = [
    {
        "role": "user",
        "content": "{}".format(text)
    }
]

prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

outputs = pipe(
    prompt,
    temperature=0.2,
    add_special_tokens=True
)
print(outputs[0]["generated_text"][len(prompt):])

results

대한민국의 수도는 서울특별시입니다.
서울특별시에는 청와대, 국회의사당, 대법원 등 대한민국의 주요 정부기관이 위치해 있습니다.
또한 서울시는 대한민국의 경제, 문화, 교육, 교통의 중심지로써 대한민국의 수도이자 대표 도시입니다.제가 도움이 되었길 바랍니다. 더 궁금한 점이 있으시면 언제든지 물어보세요!

@misc {cpm-ai/Ocelot-Ko-self-instruction-10.8B-v1.0,
    author       = { {frcp, nebchi, pepperonipizza97} },
    title        = { solar-kor-resume},
    year         = 2024,
    url          = { https://huggingface.co/cpm-ai/Ocelot-Ko-self-instruction-10.8B-v1.0 },
    publisher    = { Hugging Face }
}

Results in LogicKor* are as follows:

Model	Single turn*	Multi turn*	Overall*
gemini-1.5-pro-preview-0215	7.90	6.26	7.08
xionic-1-72b-20240404	7.23	6.28	6.76
Ocelot-Instruct	6.79	6.71	6.75
allganize/Llama-3-Alpha-Ko-8B-Instruct	7.14	6.09	6.61

Downloads last month: 15

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for cpm-ai/Llama3-Ocelot-8B-instruct-v01

Quantizations

1 model