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

Libraries

How to use AppliedIntuitionResearch/nord with Transformers:

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

pipe = pipeline("image-text-to-text", model="AppliedIntuitionResearch/nord")
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("AppliedIntuitionResearch/nord")
model = AutoModelForMultimodalLM.from_pretrained("AppliedIntuitionResearch/nord", 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 AppliedIntuitionResearch/nord with vLLM:

Install from pip and serve model

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

SGLang

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

NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning (SFT + Dr. GRPO)

CVPR 2026 | arXiv | Project Page | GitHub

Ishaan Rawal · Shubh Gupta · Yihan Hu · Wei Zhan

This is the paper's main result: Qwen2.5-VL-3B supervised fine-tuned and then further trained with Dr. GRPO to directly predict driving trajectories as discrete tokens without chain-of-thought reasoning, using 3× fewer tokens than reasoning-based VLA baselines.

Model	Training	NAVSIM PDMS
`nord` (this repo)	SFT + Dr. GRPO	0.8626
`nord-base`	SFT only	0.7273

PDMS                                 0.8626
  no_at_fault_collisions             0.9737
  drivable_area_compliance           0.9522
  ego_progress                       0.8156
  time_to_collision                  0.9253
  comfort                            0.9997

Usage

Install the nord client and serve with vLLM:

vllm serve AppliedIntuitionResearch/nord --served-model-name qwen --dtype bfloat16 --port 8000

from PIL import Image
import nord

agent = nord.NordAgent.from_pretrained("nord")

output = agent.predict(nord.NordInput(
    cameras=[Image.open("fl.jpg"), Image.open("front.jpg"), Image.open("fr.jpg")],
    ego_velocity_ms=(8.3, 0.0),
    driving_command="straight",
))

print(output.trajectory.shape)   # (40, 3) — x, y, heading at 10 Hz, 4 seconds

Full inference and NAVSIM evaluation instructions: github.com/Applied-Intuition-Open-Source/nord.

This repo bundles the K-Disc trajectory tokenizer vocab (vocab.pkl, 2048 clusters) used to decode the model's output tokens into (x, y, heading) trajectories.

Citation

@inproceedings{rawal2026nord,
  title={NoRD: A Data-Efficient Vision-Language-Action Model that Drives without Reasoning},
  author={Rawal, Ishaan and Gupta, Shubh and Hu, Yihan and Zhan, Wei},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2026}
}

License

This checkpoint is released under CC BY-NC-SA 4.0 (non-commercial). The nord inference client code is separately licensed under Apache 2.0 — see github.com/Applied-Intuition-Open-Source/nord.