Instructions to use SynLayers/Bbox-caption-8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SynLayers/Bbox-caption-8b with Transformers:

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

pipe = pipeline("image-text-to-text", model="SynLayers/Bbox-caption-8b")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("SynLayers/Bbox-caption-8b")
model = AutoModelForImageTextToText.from_pretrained("SynLayers/Bbox-caption-8b")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use SynLayers/Bbox-caption-8b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SynLayers/Bbox-caption-8b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SynLayers/Bbox-caption-8b",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/SynLayers/Bbox-caption-8b

SGLang

How to use SynLayers/Bbox-caption-8b 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 "SynLayers/Bbox-caption-8b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SynLayers/Bbox-caption-8b",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "SynLayers/Bbox-caption-8b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SynLayers/Bbox-caption-8b",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use SynLayers/Bbox-caption-8b with Docker Model Runner:
```
docker model run hf.co/SynLayers/Bbox-caption-8b
```

Bbox-caption-8b / demo /upload_used_bundle_to_hf.py

SynLayers

Upload demo/upload_used_bundle_to_hf.py with huggingface_hub

e4f1016 verified 6 days ago

raw

history blame contribute delete

5.35 kB

	from __future__ import annotations

	import argparse
	import os
	from pathlib import Path

	from huggingface_hub import HfApi

	PROJECT_ROOT = Path(__file__).resolve().parents[1]
	SHARE_ROOT = Path("/project/llmsvgen/share/data/kmw_layered_checkpoint")

	USED_FILE_MAP = {
	PROJECT_ROOT / "demo" / "model_card.md": "README.md",
	PROJECT_ROOT / "demo" / "__init__.py": "demo/__init__.py",
	PROJECT_ROOT / "demo" / "app.py": "demo/app.py",
	PROJECT_ROOT / "demo" / "hf_repo_assets.py": "demo/hf_repo_assets.py",
	PROJECT_ROOT / "demo" / "README.md": "demo/README.md",
	PROJECT_ROOT / "demo" / "publish_space.py": "demo/publish_space.py",
	PROJECT_ROOT / "demo" / "real_world_pipeline.py": "demo/real_world_pipeline.py",
	PROJECT_ROOT / "demo" / "requirements-hf-space.txt": "demo/requirements-hf-space.txt",
	PROJECT_ROOT / "demo" / "upload_used_bundle_to_hf.py": "demo/upload_used_bundle_to_hf.py",
	PROJECT_ROOT / "demo" / "infer" / "__init__.py": "demo/infer/__init__.py",
	PROJECT_ROOT / "demo" / "infer" / "run_caption_bbox_infer.py": "demo/infer/run_caption_bbox_infer.py",
	PROJECT_ROOT / "demo" / "infer" / "vlm_bbox_inference.py": "demo/infer/vlm_bbox_inference.py",
	PROJECT_ROOT / "infer" / "__init__.py": "infer/__init__.py",
	PROJECT_ROOT / "infer" / "common_infer.py": "infer/common_infer.py",
	PROJECT_ROOT / "infer" / "infer.py": "infer/infer.py",
	PROJECT_ROOT / "infer" / "infer.yaml": "infer/infer.yaml",
	PROJECT_ROOT / "models" / "__init__.py": "models/__init__.py",
	PROJECT_ROOT / "models" / "multiLayer_adapter.py": "models/multiLayer_adapter.py",
	PROJECT_ROOT / "models" / "mmdit.py": "models/mmdit.py",
	PROJECT_ROOT / "models" / "pipeline.py": "models/pipeline.py",
	PROJECT_ROOT / "models" / "transp_vae.py": "models/transp_vae.py",
	PROJECT_ROOT / "tools" / "__init__.py": "tools/__init__.py",
	PROJECT_ROOT / "tools" / "tools.py": "tools/tools.py",
	PROJECT_ROOT / "dataset_scaleup" / "dataset_construction.sh": "dataset/dataset_construction.sh",
	PROJECT_ROOT / "dataset_scaleup" / "scaleup_api.py": "dataset/scaleup_api.py",
	PROJECT_ROOT / "dataset_scaleup" / "scaleup_dataset.py": "dataset/scaleup_dataset.py",
	PROJECT_ROOT / "dataset_scaleup" / "scaleup_utils.py": "dataset/scaleup_utils.py",
	PROJECT_ROOT / "environment.yml": "environment.yml",
	}

	REQUIRED_ASSET_FILES = {
	PROJECT_ROOT / "ckpt" / "trans_vae" / "0008000.pt": "ckpt/trans_vae/0008000.pt",
	PROJECT_ROOT / "ckpt" / "pre_trained_LoRA" / "pytorch_lora_weights.safetensors": "ckpt/pre_trained_LoRA/pytorch_lora_weights.safetensors",
	PROJECT_ROOT / "ckpt" / "prism_ft_LoRA" / "pytorch_lora_weights.safetensors": "ckpt/prism_ft_LoRA/pytorch_lora_weights.safetensors",
	}

	REQUIRED_ASSET_FOLDERS = {
	SHARE_ROOT / "SynLayers_checkpoints" / "FLUX.1-dev": "SynLayers_checkpoints/FLUX.1-dev",
	SHARE_ROOT / "SynLayers_ckpt" / "step_120000": "SynLayers_ckpt/step_120000",
	SHARE_ROOT / "SynLayers_checkpoints" / "FLUX.1-dev-Controlnet-Inpainting-Alpha": "SynLayers_checkpoints/FLUX.1-dev-Controlnet-Inpainting-Alpha",
	}


	def upload_files(api: HfApi, repo_id: str, repo_type: str):
	for local_path, remote_path in USED_FILE_MAP.items():
	if not local_path.exists():
	print(f"Skipping missing file: {local_path}")
	continue
	print(f"Uploading file {local_path} -> {remote_path}")
	api.upload_file(
	path_or_fileobj=str(local_path),
	path_in_repo=remote_path,
	repo_id=repo_id,
	repo_type=repo_type,
	)

	for local_path, remote_path in REQUIRED_ASSET_FILES.items():
	if not local_path.exists():
	print(f"Skipping missing asset: {local_path}")
	continue
	print(f"Uploading asset {local_path} -> {remote_path}")
	api.upload_file(
	path_or_fileobj=str(local_path),
	path_in_repo=remote_path,
	repo_id=repo_id,
	repo_type=repo_type,
	)


	def upload_folders(api: HfApi, repo_id: str, repo_type: str):
	for local_path, remote_path in REQUIRED_ASSET_FOLDERS.items():
	if not local_path.exists():
	print(f"Skipping missing folder: {local_path}")
	continue
	print(f"Uploading folder {local_path} -> {remote_path}")
	api.upload_folder(
	folder_path=str(local_path),
	path_in_repo=remote_path,
	repo_id=repo_id,
	repo_type=repo_type,
	)


	def main():
	parser = argparse.ArgumentParser(
	description="Upload the minimal used SynLayers demo bundle to a Hugging Face repo."
	)
	parser.add_argument("--repo-id", type=str, default="SynLayers/Bbox-caption-8b")
	parser.add_argument("--repo-type", type=str, default="model", choices=["model", "space", "dataset"])
	parser.add_argument("--token", type=str, default=os.environ.get("HF_TOKEN"))
	args = parser.parse_args()

	if not args.token:
	raise ValueError("Missing Hugging Face token. Pass --token or set HF_TOKEN.")

	api = HfApi(token=args.token)
	upload_files(api, args.repo_id, args.repo_type)
	upload_folders(api, args.repo_id, args.repo_type)

	print("")
	print(f"Finished uploading used demo bundle to https://huggingface.co/{args.repo_id}")


	if __name__ == "__main__":
	main()