Text Generation
Transformers
Safetensors
English
markupdm
graphic design
design completion
multimodal
markup document
custom_code
Instructions to use cyberagent/markupdm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cyberagent/markupdm with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cyberagent/markupdm", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("cyberagent/markupdm", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use cyberagent/markupdm with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "cyberagent/markupdm" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cyberagent/markupdm", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/cyberagent/markupdm
- SGLang
How to use cyberagent/markupdm 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 "cyberagent/markupdm" \ --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": "cyberagent/markupdm", "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 "cyberagent/markupdm" \ --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": "cyberagent/markupdm", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use cyberagent/markupdm with Docker Model Runner:
docker model run hf.co/cyberagent/markupdm
Update processing_markupdm.py
Browse files- processing_markupdm.py +2 -1
processing_markupdm.py
CHANGED
|
@@ -117,7 +117,8 @@ class MarkupDMProcessor(ProcessorMixin): # type: ignore
|
|
| 117 |
example = self.preprocess_images(example["images"])
|
| 118 |
|
| 119 |
assert vision_model is not None, "Vision model must be provided."
|
| 120 |
-
image = example.pop("image")
|
|
|
|
| 121 |
with torch.inference_mode():
|
| 122 |
_, _, (_, _, image_ids) = vision_model.model.encode(image)
|
| 123 |
example["image_ids"] = list(image_ids.view(image.size(0), -1).cpu())
|
|
|
|
| 117 |
example = self.preprocess_images(example["images"])
|
| 118 |
|
| 119 |
assert vision_model is not None, "Vision model must be provided."
|
| 120 |
+
image = example.pop("image")
|
| 121 |
+
image = image.to(dtype=vision_model.dtype, device=vision_model.device)
|
| 122 |
with torch.inference_mode():
|
| 123 |
_, _, (_, _, image_ids) = vision_model.model.encode(image)
|
| 124 |
example["image_ids"] = list(image_ids.view(image.size(0), -1).cpu())
|