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
```

Document third-party inference dependencies

by OmarLaymounNvidia - opened 3 days ago

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

+26

-2

Files changed (3) hide show

batch_utils/CODE_MANIFEST.json +6 -2
batch_utils/README.md +2 -0
batch_utils/THIRD_PARTY_NOTICES.md +18 -0

batch_utils/CODE_MANIFEST.json CHANGED Viewed

@@ -1,8 +1,8 @@
 {
   "files": {
     "batch_utils/README.md": {
-      "bytes": 4125,
-      "sha256": "f9a9dd0a67b914bb1304a53674c03deec9f6a9aaa580781a13d88b365284a65c"
     },
     "batch_utils/__init__.py": {
       "bytes": 91,
@@ -167,6 +167,10 @@
     "infer.py": {
       "bytes": 171,
       "sha256": "8cddbb43ba5b9d884e8f771b9c9e219ce2e0a3c1d71a94a75d02b1d3b730f93d"
     }
   },
   "schema_version": "flora_inference_code_manifest_v1"

 {
   "files": {
     "batch_utils/README.md": {
+      "bytes": 4337,
+      "sha256": "f03e4f204b906db2b11a447585e09b6e01af5129a88b9123a5066087309f3007"
     },
     "batch_utils/__init__.py": {
       "bytes": 91,
     "infer.py": {
       "bytes": 171,
       "sha256": "8cddbb43ba5b9d884e8f771b9c9e219ce2e0a3c1d71a94a75d02b1d3b730f93d"
+    },
+    "batch_utils/THIRD_PARTY_NOTICES.md": {
+      "bytes": 1497,
+      "sha256": "b73d47aa224b69e8c97850912c23d8b29219126568a11fa41ee3d085a8a430a3"
     }
   },
   "schema_version": "flora_inference_code_manifest_v1"

batch_utils/README.md CHANGED Viewed

@@ -8,6 +8,8 @@ included; no internal source checkout or running demo is required.
 ## Setup
 Python 3.12 and a CUDA GPU with 80 GB VRAM are the validated configuration.
 The two models run sequentially. Download/cache space of at least 150 GB is
 recommended for the released FP32 Captioner, base model, and adapter.

 ## Setup
+This project will download and install additional third-party open source software projects. Review the license terms of these open source projects before use. See [third-party notices](THIRD_PARTY_NOTICES.md).
 Python 3.12 and a CUDA GPU with 80 GB VRAM are the validated configuration.
 The two models run sequentially. Download/cache space of at least 150 GB is
 recommended for the released FP32 Captioner, base model, and adapter.

batch_utils/THIRD_PARTY_NOTICES.md ADDED Viewed

	@@ -0,0 +1,18 @@

+# Third-party notices
+The inference scripts use the packages below, installed separately by the setup commands. These packages are not bundled in this repository. The requirements match [requirements.txt](requirements.txt).
+The links identify upstream project licenses. Installed distributions and their dependencies retain their own licenses and notices. For version ranges, consult the license files supplied with the installed version.
+| Requirement | Upstream license |
+| --- | --- |
+| `torch==2.11.0` | [BSD-3-Clause](https://github.com/pytorch/pytorch/blob/v2.11.0/LICENSE) |
+| `torchvision==0.26.0` | [BSD-3-Clause](https://github.com/pytorch/vision/blob/v0.26.0/LICENSE) |
+| `huggingface_hub==1.33.0` | [Apache-2.0](https://github.com/huggingface/huggingface_hub/blob/v1.33.0/LICENSE) |
+| `transformers==5.8.1` | [Apache-2.0](https://github.com/huggingface/transformers/blob/v5.8.1/LICENSE) |
+| `peft==0.18.1` | [Apache-2.0](https://github.com/huggingface/peft/blob/v0.18.1/LICENSE) |
+| `accelerate==1.12.0` | [Apache-2.0](https://github.com/huggingface/accelerate/blob/v1.12.0/LICENSE) |
+| `safetensors==0.7.0` | [Apache-2.0](https://github.com/huggingface/safetensors/blob/v0.7.0/LICENSE) |
+| `av==16.0.1` | [BSD-3-Clause](https://github.com/PyAV-Org/PyAV/blob/v16.0.1/LICENSE.txt) |
+| `numpy<3` | [BSD-3-Clause](https://github.com/numpy/numpy/blob/main/LICENSE.txt) |
+| `Pillow>=11,<13` | [PIL Software License (MIT-CMU)](https://github.com/python-pillow/Pillow/blob/main/LICENSE) |