Instructions to use xiaoyu1104/InstanceControl_canny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xiaoyu1104/InstanceControl_canny with Transformers:

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

pipe = pipeline("image-text-to-text", model="xiaoyu1104/InstanceControl_canny")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("xiaoyu1104/InstanceControl_canny", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use xiaoyu1104/InstanceControl_canny with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xiaoyu1104/InstanceControl_canny"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "xiaoyu1104/InstanceControl_canny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/xiaoyu1104/InstanceControl_canny

SGLang

How to use xiaoyu1104/InstanceControl_canny 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 "xiaoyu1104/InstanceControl_canny" \
    --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": "xiaoyu1104/InstanceControl_canny",
		"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 "xiaoyu1104/InstanceControl_canny" \
        --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": "xiaoyu1104/InstanceControl_canny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use xiaoyu1104/InstanceControl_canny with Docker Model Runner:
```
docker model run hf.co/xiaoyu1104/InstanceControl_canny
```

Add model card for Sa2va-Instance-4B (Stage 1 of InstanceControl)

by nielsr HF Staff - opened 6 days ago

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+63

-0

nielsr

6 days ago

This PR adds a comprehensive model card for Sa2va-Instance-4B, which is the Stage 1 model of the InstanceControl framework presented in the ECCV 2026 paper InstanceControl: Controllable Complex Image Generation without Instance Labeling.

The model card includes:

Metadata specifying the image-text-to-text pipeline tag and the transformers library name.
References and links to the paper, project page, GitHub repository, and dataset/benchmark.
Installation instructions and usage examples for performing Stage 1 inference to predict instance masks.

Add model card for Sa2va-Instance-4B (Stage 1 of InstanceControl)dbd7e890

xiaoyu1104 changed pull request status to merged 4 days ago

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