Instructions to use SeanScripts/pixtral-12b-nf4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SeanScripts/pixtral-12b-nf4 with Transformers:

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

pipe = pipeline("image-text-to-text", model="SeanScripts/pixtral-12b-nf4")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("SeanScripts/pixtral-12b-nf4")
model = AutoModelForImageTextToText.from_pretrained("SeanScripts/pixtral-12b-nf4")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use SeanScripts/pixtral-12b-nf4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SeanScripts/pixtral-12b-nf4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SeanScripts/pixtral-12b-nf4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/SeanScripts/pixtral-12b-nf4

SGLang

How to use SeanScripts/pixtral-12b-nf4 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 "SeanScripts/pixtral-12b-nf4" \
    --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": "SeanScripts/pixtral-12b-nf4",
		"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 "SeanScripts/pixtral-12b-nf4" \
        --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": "SeanScripts/pixtral-12b-nf4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use SeanScripts/pixtral-12b-nf4 with Docker Model Runner:
```
docker model run hf.co/SeanScripts/pixtral-12b-nf4
```

working python packages

by MJ-12 - opened Oct 2, 2024

Discussion

MJ-12

Oct 2, 2024

I got this working with the following python packages:
accelerate==0.34.2
bitsandbytes==0.44.1
certifi==2024.8.30
charset-normalizer==3.3.2
colorama==0.4.6
filelock==3.16.1
fsspec==2024.9.0
huggingface-hub==0.25.1
idna==3.10
Jinja2==3.1.3
MarkupSafe==2.1.5
mpmath==1.3.0
networkx==3.2.1
numpy==2.1.1
packaging==24.1
pillow==10.2.0
psutil==6.0.0
PyYAML==6.0.2
regex==2024.9.11
requests==2.32.3
safetensors==0.4.5
sympy==1.12
tokenizers==0.20.0
torch==2.4.1+cu121
torchvision==0.19.1+cu121
tqdm==4.66.5
transformers==4.45.1
typing_extensions==4.12.2
urllib3==2.2.3

fusi0n

Oct 7, 2024

•

edited Oct 7, 2024

going from 4.40 to 4.45.1 with transformers fixed my issues.

5t/s on a 4060ti 16gb.

MJ-12

Oct 8, 2024

@fusi0n same card here, how did you find the outputs? I found the inference to be totally hit and miss though. I think a bigger parameter model is needed.

fusi0n

Oct 17, 2024

@info-int output was not great, i perfer moondream2 for my use case, but it sucks losing the true multimodel aspect

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