Instructions to use meta-llama/CodeLlama-7b-Instruct-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meta-llama/CodeLlama-7b-Instruct-hf with Transformers:

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

pipe = pipeline("text-generation", model="meta-llama/CodeLlama-7b-Instruct-hf")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("meta-llama/CodeLlama-7b-Instruct-hf")
model = AutoModelForCausalLM.from_pretrained("meta-llama/CodeLlama-7b-Instruct-hf")
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 meta-llama/CodeLlama-7b-Instruct-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "meta-llama/CodeLlama-7b-Instruct-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-llama/CodeLlama-7b-Instruct-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/meta-llama/CodeLlama-7b-Instruct-hf

SGLang

How to use meta-llama/CodeLlama-7b-Instruct-hf 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 "meta-llama/CodeLlama-7b-Instruct-hf" \
    --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": "meta-llama/CodeLlama-7b-Instruct-hf",
		"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 "meta-llama/CodeLlama-7b-Instruct-hf" \
        --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": "meta-llama/CodeLlama-7b-Instruct-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use meta-llama/CodeLlama-7b-Instruct-hf with Docker Model Runner:
```
docker model run hf.co/meta-llama/CodeLlama-7b-Instruct-hf
```

How to prompt it?

by rakotomandimby - opened Oct 5, 2024

Discussion

rakotomandimby

Oct 5, 2024

•

edited Oct 5, 2024

Hello, I have the following simple Python script:

from transformers import AutoTokenizer
import transformers
import torch

model = "meta-llama/CodeLlama-7b-Instruct-hf"
tokenizer = AutoTokenizer.from_pretrained(model) 
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

sequences = pipeline(
    'Write Python code using the model "'+model+'",  AutoTokenizer, transformer and torch Python modules that will start an HTTP server and take the prompt from the body of a POST request. The result will be sent as response.',
    do_sample=True,
    top_k=10,
    temperature=0.1,
    top_p=0.95,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
    max_length=512,
    truncation=True)
for seq in sequences:
    print(f"Result: {seq['generated_text']}")

The only result I get is:

### Prerequisites

* Python 3.6+
* Transformers 4.12.4
* Torch 1.10.0
* AutoTokenizer 0.11.0

### Installing


pip install -r requirements.txt


### Running the server


python server.py


### Testing the server


curl -X POST -H "Content-Type: application/json" -d '{"prompt": "What is the capital of France?"}' http://localhost:8000/


### Built With

* [Transformers](https://github.com/huggingface/transformers) - The library used to implement the model
* [Torch](https://github.com/pytorch/pytorch) - The library used to implement the model
* [AutoTokenizer](https://github.com/huggingface/tokenizers) - The library used to implement the tokenizer

### Authors

* **Thomas BERNARD** - *Initial work* - [thomasbernard](https://github.com/thomasbernard)

### License

This project is licensed under the MIT License - see the [LICENSE.md](LICENSE.md) file for details

I guess I miss some docs on how to prompt it. Could you help me by pointing me to some tutorials?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment