Instructions to use togethercomputer/LLaMA-2-7B-32K with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use togethercomputer/LLaMA-2-7B-32K with Transformers:

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

pipe = pipeline("text-generation", model="togethercomputer/LLaMA-2-7B-32K")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("togethercomputer/LLaMA-2-7B-32K")
model = AutoModelForCausalLM.from_pretrained("togethercomputer/LLaMA-2-7B-32K")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use togethercomputer/LLaMA-2-7B-32K with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "togethercomputer/LLaMA-2-7B-32K"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "togethercomputer/LLaMA-2-7B-32K",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/togethercomputer/LLaMA-2-7B-32K

SGLang

How to use togethercomputer/LLaMA-2-7B-32K 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 "togethercomputer/LLaMA-2-7B-32K" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "togethercomputer/LLaMA-2-7B-32K",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "togethercomputer/LLaMA-2-7B-32K" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "togethercomputer/LLaMA-2-7B-32K",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use togethercomputer/LLaMA-2-7B-32K with Docker Model Runner:
```
docker model run hf.co/togethercomputer/LLaMA-2-7B-32K
```

What is the VRAM requirement of this model?

by Said2k - opened Jul 29, 2023

Discussion

Said2k

Jul 29, 2023

What is the VRAM requirement of this model? I have 8 GB VRAM and I was wondering if this model could be run on that much?

ThatOneShortGuy

Jul 29, 2023

If you have bitsandbytes, you should be able to load the model with load_in_8bit=Trueparam in your AutoModelForCausalLM func

zhangce

Together org Jul 29, 2023

I don't think VRAM 8GB is enough for this unfortunately (especially given that when we go to 32K, the size of KV cache becomes quite large too) -- we are pushing to decrease this! (e.g., we could do some KV cache quantization similar to what we have done in https://arxiv.org/abs/2303.06865, but it will take time)

In the meantime, you can go to https://api.together.xyz/playground to play with it!

BajrangWappnet

Aug 3, 2023

How can we load the model using bitsandbytes ?

mauriceweber

Aug 4, 2023

@BajrangWappnet , I think you can just do something like this:

model = AutoModelForCausalLM.from_pretrained(
    "togethercomputer/LLaMA-2-7B-32K", 
    trust_remote_code=False, 
    torch_dtype=torch.float16,
    load_in_8bit=True
)

Here's a more detailed example on how to use bitsandbytes: https://github.com/TimDettmers/bitsandbytes/blob/main/examples/int8_inference_huggingface.py

MathewOpt

Aug 15, 2023

This comment has been hidden

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