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
```

SynLayers commited on 7 days ago

Commit

0203586

verified ·

1 Parent(s): e763920

Upload demo/README.md with huggingface_hub

Browse files

Files changed (1) hide show

demo/README.md +98 -0

demo/README.md ADDED Viewed

	@@ -0,0 +1,98 @@

+# SynLayers Demo
+This folder now contains a unified real-world inference demo:
+1. `demo/infer` runs the fixed-prompt VLM caption + bbox detector.
+2. `infer/infer.py` runs SynLayers decomposition with `infer/infer.yaml`.
+3. `demo/real_world_pipeline.py` stitches the two stages together for one uploaded image.
+4. `demo/app.py` provides a Gradio interface that can be used locally or adapted for a Hugging Face Space.
+5. `demo/upload_used_bundle_to_hf.py` uploads only the Python/config files actually used by the demo, plus the selected runtime assets.
+## Local Run
+From the `SynLayers` root:
+```bash
+python demo/app.py
+```
+Or run the unified CLI directly:
+```bash
+python demo/real_world_pipeline.py \
+  --image "/path/to/your/image.png"
+```
+## Default Models
+The current local defaults are:
+- bbox-caption model:
+  `/project/llmsvgen/share/data/kmw_layered_checkpoint/Bbox-caption-8b`
+- SynLayers base checkpoints:
+  `/project/llmsvgen/share/data/kmw_layered_checkpoint/SynLayers_checkpoints`
+- SynLayers decomposition checkpoint:
+  `/project/llmsvgen/share/data/kmw_layered_checkpoint/SynLayers_ckpt/step_120000`
+- base config:
+  `infer/infer.yaml`
+## Hugging Face Space Notes
+The Gradio app is ready, but the current default model paths are cluster-local absolute paths.
+That means an online Hugging Face Space will only run after the model assets are made available
+inside the Space runtime.
+The app supports overriding the local defaults with environment variables:
+- `SYNLAYERS_BBOX_MODEL`
+- `SYNLAYERS_BASE_MODEL`
+- `SYNLAYERS_ADAPTER_MODEL`
+- `SYNLAYERS_TRANSP_VAE`
+- `SYNLAYERS_PRETRAINED_LORA`
+- `SYNLAYERS_ARTPLUS_LORA`
+- `SYNLAYERS_DECOMP_CKPT_ROOT`
+- `SYNLAYERS_REAL_CONFIG`
+- `SYNLAYERS_DEMO_WORK_DIR`
+- `SYNLAYERS_EXAMPLE_DIR`
+In practice, for a real Hugging Face Space deployment you will want to:
+1. upload or mirror the required model assets to Hugging Face-accessible storage
+2. set the environment variables above in the Space settings
+3. launch `demo/app.py` as the Space entrypoint
+## Upload Bundle
+To upload the minimal used demo bundle to a Hugging Face repo:
+```bash
+python demo/upload_used_bundle_to_hf.py \
+  --repo-id SynLayers/Bbox-caption-8b
+```
+This uploads:
+- the used `demo`, `infer`, `models`, and `tools` Python files
+- `infer/infer.yaml`
+- `environment.yml`
+- `ckpt/trans_vae/0008000.pt`
+- `ckpt/pre_trained_LoRA/pytorch_lora_weights.safetensors`
+- `ckpt/prism_ft_LoRA/pytorch_lora_weights.safetensors`
+- `SynLayers_ckpt/step_120000`
+- `SynLayers_checkpoints/FLUX.1-dev-Controlnet-Inpainting-Alpha`
+If you also want to upload the large base checkpoints, add:
+```bash
+python demo/upload_used_bundle_to_hf.py \
+  --repo-id SynLayers/Bbox-caption-8b \
+  --include-base-checkpoints
+```
+## Fixed Prompt
+The bbox detector always uses the fixed prompt defined in:
+- `demo/infer/run_caption_bbox_infer.py`
+No extra user text prompt is required.