Instructions to use edbeeching/vsft-idefics2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use edbeeching/vsft-idefics2 with Transformers:

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

pipe = pipeline("image-text-to-text", model="edbeeching/vsft-idefics2")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("edbeeching/vsft-idefics2")
model = AutoModelForImageTextToText.from_pretrained("edbeeching/vsft-idefics2")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use edbeeching/vsft-idefics2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "edbeeching/vsft-idefics2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "edbeeching/vsft-idefics2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/edbeeching/vsft-idefics2

SGLang

How to use edbeeching/vsft-idefics2 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 "edbeeching/vsft-idefics2" \
    --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": "edbeeching/vsft-idefics2",
		"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 "edbeeching/vsft-idefics2" \
        --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": "edbeeching/vsft-idefics2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use edbeeching/vsft-idefics2 with Docker Model Runner:
```
docker model run hf.co/edbeeching/vsft-idefics2
```

edbeeching HF Staff commited on Apr 16, 2024

Commit

1e2a841

verified ·

1 Parent(s): c6e5990

Upload 2 files

Browse files

Files changed (2) hide show

preprocessor_config.json +26 -0
processor_config.json +4 -0

preprocessor_config.json ADDED Viewed

	@@ -0,0 +1,26 @@

+{
+  "do_convert_rgb": true,
+  "do_image_splitting": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "Idefics2ImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "processor_class": "Idefics2Processor",
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 980,
+    "shortest_edge": 378
+  }
+}

processor_config.json ADDED Viewed

	@@ -0,0 +1,4 @@

+{
+  "image_seq_len": 64,
+  "processor_class": "Idefics2Processor"
+}