Instructions to use google/paligemma-3b-pt-896 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/paligemma-3b-pt-896 with Transformers:

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

pipe = pipeline("image-text-to-text", model="google/paligemma-3b-pt-896")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("google/paligemma-3b-pt-896")
model = AutoModelForImageTextToText.from_pretrained("google/paligemma-3b-pt-896")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use google/paligemma-3b-pt-896 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "google/paligemma-3b-pt-896"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "google/paligemma-3b-pt-896",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/google/paligemma-3b-pt-896

SGLang

How to use google/paligemma-3b-pt-896 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 "google/paligemma-3b-pt-896" \
    --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": "google/paligemma-3b-pt-896",
		"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 "google/paligemma-3b-pt-896" \
        --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": "google/paligemma-3b-pt-896",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use google/paligemma-3b-pt-896 with Docker Model Runner:
```
docker model run hf.co/google/paligemma-3b-pt-896
```

Training version 896

by lcolonn - opened Aug 8, 2024

Discussion

lcolonn

Aug 8, 2024

Has anyone managed to train this model yet? Due to the large image token sequence length, it requires using a sharding strategy. I setup all my code with Lightning rather than HF and accelerate (which was probably a mistake) and am still unable to get it to train - I am running into errors with FSDP. Wondering if anyone managed to train the model.

lcolonn

Aug 17, 2024

•

edited Aug 17, 2024

If this is of any help to others, using Lightning and FSDP with 8 A100-80GB I am now able to train it with a batch size of 2 in fp32. I'll try to release my fine-tuning code soon. They key here is the activation checkpointing as it's the activations taking up the vast majority of VRAM due to the large sequence length.

kwin-sustainment

Aug 23, 2024

@lcolonn did you ever release that fine-tuning code?

lcolonn

Aug 27, 2024

Hi @kwin-sustainment , I'm quite busy finishing a project at the moment and was planning a release in around two weeks time. However, if it's urgent for you, I can send over the main elements

kwin-sustainment

Aug 27, 2024

•

edited Aug 27, 2024

@lcolonn hey no worries! I actually got it working after I delved into your hint about that checkpointing & VRAM. Thank you for the advice :)

Edit: I think you need a MINIMUM of a 24GB card to train this model for people looking at this thread in the future. It would be a batch size of one and take forever, but it'd work...

lkv

Google org Oct 22, 2024

Hi @kwin-sustainment , I hope the issue has been resolved feel free to close this issue . Please let us know if any further assistance is needed. 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