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

Libraries

How to use NeuralMetrics/RolmOCR with Transformers:

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

pipe = pipeline("image-text-to-text", model="NeuralMetrics/RolmOCR")
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, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("NeuralMetrics/RolmOCR")
model = AutoModelForMultimodalLM.from_pretrained("NeuralMetrics/RolmOCR", device_map="auto")
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 Settings

vLLM

How to use NeuralMetrics/RolmOCR with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NeuralMetrics/RolmOCR"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NeuralMetrics/RolmOCR",
		"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/NeuralMetrics/RolmOCR

SGLang

How to use NeuralMetrics/RolmOCR 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 "NeuralMetrics/RolmOCR" \
    --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": "NeuralMetrics/RolmOCR",
		"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 "NeuralMetrics/RolmOCR" \
        --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": "NeuralMetrics/RolmOCR",
		"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 NeuralMetrics/RolmOCR with Docker Model Runner:
```
docker model run hf.co/NeuralMetrics/RolmOCR
```

mkvn commited on about 18 hours ago

Commit

3e726f2

verified ·

1 Parent(s): 9d49496

Add Neural Metrics card (upstream card preserved below)

Browse files

Files changed (1) hide show

README.md +26 -1

README.md CHANGED Viewed

@@ -5,6 +5,29 @@ datasets:
 - allenai/olmOCR-mix-0225
 base_model: Qwen/Qwen2.5-VL-7B-Instruct
 ---
 # RolmOCR by [Reducto AI](https://reducto.ai/)
@@ -89,4 +112,6 @@ print(ocr_page_with_rolm(img_base64))
   title = {RolmOCR: A Faster, Lighter Open Source OCR Model},
   year = {2025},
 }
-```

 - allenai/olmOCR-mix-0225
 base_model: Qwen/Qwen2.5-VL-7B-Instruct
 ---
+<div align="center">
+# Neural Metrics · A faster, lighter take on the olmOCR recipe.
+<img src="https://img.shields.io/badge/Neural%20Metrics-document%20extraction-4F46E5?style=for-the-badge" alt="Neural Metrics" />
+<img src="https://img.shields.io/badge/fork%20of-reducto%2FRolmOCR-2563EB?style=flat-square" alt="fork" />
+</div>
+Reducto's RolmOCR builds on the olmOCR approach with a Qwen2.5-VL backbone, trading some setup complexity for speed and lower memory use - no PDF metadata prompting required.
+**We use it for:** throughput-sensitive batch OCR - head-to-head comparison against olmOCR on our own document mix.
+> ### Attribution
+> This is an **unmodified fork** of [`reducto/RolmOCR`](https://huggingface.co/reducto/RolmOCR), created by the [Qwen team](https://huggingface.co/Qwen).
+> All weights, files and behaviour are identical to upstream — we rehost it so our experiments stay
+> reproducible and version-pinned. The original license and all credit remain with the Qwen team.
+> If you want the canonical model, please use [the original](https://huggingface.co/reducto/RolmOCR).
+---
+<details>
+<summary><b>Original model card from reducto/RolmOCR</b> (click to expand)</summary>
 # RolmOCR by [Reducto AI](https://reducto.ai/)
   title = {RolmOCR: A Faster, Lighter Open Source OCR Model},
   year = {2025},
 }
+```
+</details>