Instructions to use liuhaotian/llava-llama-2-13b-chat-lightning-preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use liuhaotian/llava-llama-2-13b-chat-lightning-preview with Transformers:

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

pipe = pipeline("text-generation", model="liuhaotian/llava-llama-2-13b-chat-lightning-preview")

# Load model directly
from transformers import AutoProcessor, AutoModelForCausalLM

processor = AutoProcessor.from_pretrained("liuhaotian/llava-llama-2-13b-chat-lightning-preview")
model = AutoModelForCausalLM.from_pretrained("liuhaotian/llava-llama-2-13b-chat-lightning-preview", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use liuhaotian/llava-llama-2-13b-chat-lightning-preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "liuhaotian/llava-llama-2-13b-chat-lightning-preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "liuhaotian/llava-llama-2-13b-chat-lightning-preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/liuhaotian/llava-llama-2-13b-chat-lightning-preview

SGLang

How to use liuhaotian/llava-llama-2-13b-chat-lightning-preview 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 "liuhaotian/llava-llama-2-13b-chat-lightning-preview" \
    --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": "liuhaotian/llava-llama-2-13b-chat-lightning-preview",
		"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 "liuhaotian/llava-llama-2-13b-chat-lightning-preview" \
        --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": "liuhaotian/llava-llama-2-13b-chat-lightning-preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use liuhaotian/llava-llama-2-13b-chat-lightning-preview with Docker Model Runner:
```
docker model run hf.co/liuhaotian/llava-llama-2-13b-chat-lightning-preview
```

Why is it text generation pipeline?

by usamacogniz - opened Aug 29, 2023

Discussion

usamacogniz

Aug 29, 2023

Brother I love your work.
I have a question, why did you select text generation pipeline for all these llava models? Why didnt you select Visual Question Answering Multimodal pipeline?
Dont you think that will make the usage with Huggingface API much easier for these Llava models?

liuhaotian

Owner Sep 1, 2023

Thank you for the suggestion. The pipeline is automatically selected by HF (we did not choose a pipeline).
I am not very familiar with the HF API and inference endpoint, and we are working to improve the integration. Contribution is welcomed! Thanks.

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