Instructions to use google/paligemma-3b-pt-448 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/paligemma-3b-pt-448 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-448")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("google/paligemma-3b-pt-448")
model = AutoModelForImageTextToText.from_pretrained("google/paligemma-3b-pt-448")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use google/paligemma-3b-pt-448 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-448"
# 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-448",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/google/paligemma-3b-pt-448

SGLang

How to use google/paligemma-3b-pt-448 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-448" \
    --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-448",
		"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-448" \
        --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-448",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use google/paligemma-3b-pt-448 with Docker Model Runner:
```
docker model run hf.co/google/paligemma-3b-pt-448
```

Manual training

by edmond - opened May 22, 2024

Discussion

edmond

May 22, 2024

Hello, for some reason running :
vlm = PaliGemmaForConditionalGeneration.from_pretrained(**llm_args)
pred = vlm(pixel_values=tensor, input_ids=input_ids[:, :-1],
attention_mask=torch.ones_like(input_ids[:, :-1])).logits
pred = pred[:, -nb_tokens_answer:]

loss = F.cross_entropy(pred.permute((0, 2, 1)), input_ids[:, -nb_tokens_answer:],
reduction='mean')

Gives me a very small loss. I have the feeling that input and target tokens were mixed.
Why is that ?

edmond

May 23, 2024

•

edited May 23, 2024

This is driving me crazy. This bugfix was supposed to solve my problem https://github.com/huggingface/transformers/pull/30967 ... (im checking on more data)

edmond

May 24, 2024

•

edited May 24, 2024

https://github.com/huggingface/transformers/issues/30993 ok I got help, apparently this models neededs also labels tokens and tokens type ids in input unlike imp or moondream...

edmond changed discussion status to closed May 24, 2024

merve

May 27, 2024

@edmond sorry for late response. it's best if you pass suffix to processor and actually pass processor outputs.

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