Instructions to use nvidia/Flora-Captioner with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/Flora-Captioner with Transformers:

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

pipe = pipeline("image-text-to-text", model="nvidia/Flora-Captioner")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("nvidia/Flora-Captioner")
model = AutoModelForMultimodalLM.from_pretrained("nvidia/Flora-Captioner", 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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nvidia/Flora-Captioner with vLLM:

Install from pip and serve model

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

SGLang

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

Related work and citation request: FineVLA (arXiv:2605.27284)

by EricsXt - opened 14 days ago

Discussion

EricsXt

14 days ago

Hi Flora team,

Thanks for releasing Flora-Captioner and the broader Flora model suite. This looks like a solid contribution: making a robotics-specialized 27B captioner publicly available is valuable for researchers building scalable robot-data annotation pipelines.

I’m the first author of FineVLA, and I wanted to bring a closely related earlier work to your attention:

Hu et al., “FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies.” First released on May 26, 2026.
Paper: https://arxiv.org/abs/2605.27284

FineVLA introduces RoboFine-VLM, a Qwen3.5-based VLM fine-tuned as a scalable robot-video annotator. It generates step-level descriptions across ten fine-grained dimensions, including the active arm, contact and approach, trajectory and orientation, and failure and recovery. The paper also shows that mixing human-verified fine-grained instructions with coarse task labels improves VLA performance and steerability in simulation and real-world bimanual manipulation.

The connection is particularly relevant to Flora’s contextual action-captioning component: both use fine-tuned Qwen3.5 VLMs to turn robot videos into action-level language annotations for robotics data pipelines. Their annotation objectives differ: Flora-Captioner produces concise intended-action instructions for supplied temporal intervals, while RoboFine-VLM emphasizes detailed descriptions of how actions are executed.

Given this overlap, I believe FineVLA is relevant prior work. I could not find it cited in the current Flora paper or model card, and I would appreciate it if you could add a citation and briefly discuss the relationship between the two approaches.

Happy to share further details or compare annotation schemas if helpful. Thanks again for making these models available!

Best,
Xintong Hu

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment