Instructions to use gradientai/Llama-3-8B-Instruct-Gradient-1048k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use gradientai/Llama-3-8B-Instruct-Gradient-1048k with Transformers:

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

pipe = pipeline("text-generation", model="gradientai/Llama-3-8B-Instruct-Gradient-1048k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("gradientai/Llama-3-8B-Instruct-Gradient-1048k")
model = AutoModelForCausalLM.from_pretrained("gradientai/Llama-3-8B-Instruct-Gradient-1048k")
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 gradientai/Llama-3-8B-Instruct-Gradient-1048k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gradientai/Llama-3-8B-Instruct-Gradient-1048k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "gradientai/Llama-3-8B-Instruct-Gradient-1048k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/gradientai/Llama-3-8B-Instruct-Gradient-1048k

SGLang

How to use gradientai/Llama-3-8B-Instruct-Gradient-1048k 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 "gradientai/Llama-3-8B-Instruct-Gradient-1048k" \
    --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": "gradientai/Llama-3-8B-Instruct-Gradient-1048k",
		"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 "gradientai/Llama-3-8B-Instruct-Gradient-1048k" \
        --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": "gradientai/Llama-3-8B-Instruct-Gradient-1048k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use gradientai/Llama-3-8B-Instruct-Gradient-1048k with Docker Model Runner:
```
docker model run hf.co/gradientai/Llama-3-8B-Instruct-Gradient-1048k
```

Memory requirements to take advantage of full context window

#23

by andrewrreed - opened May 20, 2024

Discussion

andrewrreed

May 20, 2024

Thanks for your awesome work on this! What is the recommended hardware setup / inference server configs for running this model while taking advantage of full 1M context window?

My understanding is that since this model was fine-tuned with step-wise RoPE theta scaling, we wouldn't actually apply any RoPE scaling args for inference, right? Doesn't that come with the drawback of huge VRAM requirement to actually use the model with these long sequence lengths? Like >1TB of VRAM needed for 262k... making the model somewhat unusable for the gpu-poor?

Seems like dynamic RoPE scaling on the base Llama3-8B model alone would be more resource efficient for achieving really long contexts... though I guess with the tradeoff of lower quality at those long contexts....

Have you ablated the long context performance against this baseline, dynamic RoPE scaling scenario? Thanks in advance for you thoughts on this!

CS437help

Jun 1, 2024

Probably yes it requires lots of VRAM, when I used 262k, I tried with different context length while running A100 with 40 gb memory and I was only capable to give 125k input and 39 gb of memory was used during inference.

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