Instructions to use TheBloke/llava-v1.5-13B-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/llava-v1.5-13B-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="TheBloke/llava-v1.5-13B-GPTQ")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TheBloke/llava-v1.5-13B-GPTQ")
model = AutoModelForCausalLM.from_pretrained("TheBloke/llava-v1.5-13B-GPTQ", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use TheBloke/llava-v1.5-13B-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBloke/llava-v1.5-13B-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/llava-v1.5-13B-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TheBloke/llava-v1.5-13B-GPTQ

SGLang

How to use TheBloke/llava-v1.5-13B-GPTQ 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 "TheBloke/llava-v1.5-13B-GPTQ" \
    --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": "TheBloke/llava-v1.5-13B-GPTQ",
		"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 "TheBloke/llava-v1.5-13B-GPTQ" \
        --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": "TheBloke/llava-v1.5-13B-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TheBloke/llava-v1.5-13B-GPTQ with Docker Model Runner:
```
docker model run hf.co/TheBloke/llava-v1.5-13B-GPTQ
```

Model produces arbitrary output

by rene-hof - opened Jan 21, 2024

Discussion

rene-hof

Jan 21, 2024

I've collected a dataset with celebrities and ran it with different models.
It runs fine with "llava-hf/llava-1.5-7b-hf", instructBlip and CogVLM.
However, with this model, the output is just arbitrary. If you set temperature to 0.1 it always outputs Abraham Lincoln, no matter what image.
If you set it to 0.7 as suggested in the model card, it varies a lot, but there is no connection between the output and the pictures. Not even woman and man are correct, nor black and white people. Something went completely wrong here.

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