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

Libraries

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

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

pipe = pipeline("text-generation", model="codellama/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("codellama/CodeLlama-7b-Instruct-hf")
model = AutoModelForCausalLM.from_pretrained("codellama/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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

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

Use Docker

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

SGLang

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

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

Context Length and GPU VRAM Usage in CodeLlama-7B

#26

by humza-sami - opened Jan 31, 2024

Discussion

humza-sami

Jan 31, 2024

I am currently using CodeLlama-7B on an RTX 3090 24GB GPU, and I have a question regarding the relationship between context length and VRAM usage. According to the model documentation, the context length of CodeLlama-7B is 16,384 tokens.

I loaded the model using Hugging Face with 8-bit precision as follows:

from transformers import AutoModelForCausalLM, AutoTokenizer

agent_name = "codellama/CodeLlama-7b-Instruct-hf"
agent = AutoModelForCausalLM.from_pretrained(agent_name, device_map='cuda', load_in_8bit=True)
agent_tokenizer = AutoTokenizer.from_pretrained(agent_name, add_special_tokens=False, add_eos_token=False, add_bos_token=False)

I then tested the model with different input lengths. For a 3000-token input, the GPU VRAM usage was 16GB. However, when I provided a 6000-token input, the GPU VRAM spiked to 22GB. My primary concern is understanding the relationship between context length and VRAM usage.

Code for Reference:

text = 6000 * "hello "
encoded_input = agent_tokenizer(text, return_tensors="pt").to("cuda")
response = agent.generate(**encoded_input, max_new_tokens=4000, do_sample=True, temperature=0.25)

Questions:

Is my understanding correct that the model can handle inputs up to its context length of 16,384 tokens?
Could you provide insights into the observed increase in VRAM usage from a 3000-token input to a 6000-token input?
Considering my intention to use CodeLlama-7B with a context length of up to 8000 tokens, would the 24GB VRAM of my RTX 3090 be sufficient?

Any clarification on these matters would be greatly appreciated. Thank you!

ArthurZ

Code Llama org Feb 2, 2024

An easy relationship is the amount of past_key_values that you need to keep track of. The longer the sequence, the more context you have, the bigger the cache and the bigger the RAM usage !

humza-sami changed discussion status to closed Feb 2, 2024

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