Instructions to use Dharma-AI/Dharma-OCR-LITE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Dharma-AI/Dharma-OCR-LITE with Transformers:

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

pipe = pipeline("image-text-to-text", model="Dharma-AI/Dharma-OCR-LITE")
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("Dharma-AI/Dharma-OCR-LITE")
model = AutoModelForMultimodalLM.from_pretrained("Dharma-AI/Dharma-OCR-LITE", 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 Dharma-AI/Dharma-OCR-LITE with vLLM:

Install from pip and serve model

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

SGLang

How to use Dharma-AI/Dharma-OCR-LITE 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 "Dharma-AI/Dharma-OCR-LITE" \
    --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": "Dharma-AI/Dharma-OCR-LITE",
		"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 "Dharma-AI/Dharma-OCR-LITE" \
        --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": "Dharma-AI/Dharma-OCR-LITE",
		"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 Dharma-AI/Dharma-OCR-LITE with Docker Model Runner:
```
docker model run hf.co/Dharma-AI/Dharma-OCR-LITE
```

Dharma-OCR-LITE

Commit History

Apr 17

12:10
Update README.md

GabrielPimenta99

Verified signature
3fd38b0

Apr 15

12:32
Update README.md

GabrielPimenta99

Verified signature
b8326fb

Apr 14

19:03
Update README.md

GabrielPimenta99

Verified signature
80fe073
19:00
Update README.md

GabrielPimenta99

Verified signature
cc0e4de
18:59
Update README.md

GabrielPimenta99

Verified signature
ae7b3cd
18:48
Update README.md

GabrielPimenta99

Verified signature
143f8a1
18:42
Upload NOTICE.txt

GabrielPimenta99

Verified signature
1a1e9da
18:41
Update LICENSE

GabrielPimenta99

Verified signature
86471c0
18:40
Update README.md

GabrielPimenta99

Verified signature
b7d02e2
18:36
Update README.md

GabrielPimenta99

Verified signature
b33d8c1

Apr 2

17:48
Update README.md

GabrielPimenta99

Verified signature
1022571
17:44
Update README.md

GabrielPimenta99

Verified signature
aa63374

Mar 31

00:51
Update README.md

GabrielPimenta99

Verified signature
c8bfeb7

Mar 30

20:15
Move model files to repository root

Paulo-DharmaAI

Verified signature
8d42fd5
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/vocab.json with huggingface_hub

Paulo-DharmaAI

Verified signature
a2b7c8b
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/video_preprocessor_config.json with huggingface_hub

Paulo-DharmaAI

Verified signature
2c819fc
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/tokenizer_config.json with huggingface_hub

Paulo-DharmaAI

Verified signature
360e39c
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/tokenizer.json with huggingface_hub

Paulo-DharmaAI

Verified signature
cf47d0c
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/special_tokens_map.json with huggingface_hub

Paulo-DharmaAI

Verified signature
ffd3e0c
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/recipe.yaml with huggingface_hub

Paulo-DharmaAI

Verified signature
27ebf67
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/preprocessor_config.json with huggingface_hub

Paulo-DharmaAI

Verified signature
9f8ae30
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/model.safetensors.index.json with huggingface_hub

Paulo-DharmaAI

Verified signature
55fa141
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/model-00002-of-00002.safetensors with huggingface_hub

Paulo-DharmaAI

Verified signature
ced6674
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/model-00001-of-00002.safetensors with huggingface_hub

Paulo-DharmaAI

Verified signature
a8425c9
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/merges.txt with huggingface_hub

Paulo-DharmaAI

Verified signature
fc827ce
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/generation_config.json with huggingface_hub

Paulo-DharmaAI

Verified signature
c5763e0
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/config.json with huggingface_hub

Paulo-DharmaAI

Verified signature
c006185
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/chat_template.jinja with huggingface_hub

Paulo-DharmaAI

Verified signature
991abce
20:14
Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/added_tokens.json with huggingface_hub

Paulo-DharmaAI

Verified signature
27542a6
20:08
Upload folder using huggingface_hub

Paulo-DharmaAI

Verified signature
8d4f964
18:33
Update README.md

GabrielPimenta99

Verified signature
1a84827
18:29
Update README.md

GabrielPimenta99

Verified signature
da5bdd6
18:24
Update README.md

GabrielPimenta99

Verified signature
71fe0ff
18:17
Update README.md

GabrielPimenta99

Verified signature
7f7b946
18:11
Update README.md

GabrielPimenta99

Verified signature
a2c4f97
18:05
Update README.md

GabrielPimenta99

Verified signature
da942ba
17:58
Update README.md

GabrielPimenta99

Verified signature
ea434dd
17:56
Update README.md

GabrielPimenta99

Verified signature
3e9ee3e
17:47
Upload 2 files

GabrielPimenta99

Verified signature
08a76c1
14:37
Update README.md

GabrielPimenta99

Verified signature
c32f73a
14:35
Update README.md

GabrielPimenta99

Verified signature
4a4f8d6
14:31
Update README.md

GabrielPimenta99

Verified signature
a89537c
14:24
Update README.md

GabrielPimenta99

Verified signature
f922957
14:23
Update README.md

GabrielPimenta99

Verified signature
4cbe11f
14:21
Update README.md

GabrielPimenta99

Verified signature
a1af1a5
14:21
Update README.md

GabrielPimenta99

Verified signature
6fab8ec
14:19
Update README.md

GabrielPimenta99

Verified signature
93225ce
14:15
Update README.md

GabrielPimenta99

Verified signature
fec5732
14:14
Delete images/logo_horizontal_black_whiteback.png

GabrielPimenta99

Verified signature
f751cd3
14:14
Delete images/logo_dharma_ai.png

GabrielPimenta99

Verified signature
b1cd7a4

Commit History

Apr 17

Update README.md

Apr 15

Update README.md

Apr 14

Update README.md

Update README.md

Update README.md

Update README.md

Upload NOTICE.txt

Update LICENSE

Update README.md

Update README.md

Apr 2

Update README.md

Update README.md

Mar 31

Update README.md

Mar 30

Move model files to repository root

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/vocab.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/video_preprocessor_config.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/tokenizer_config.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/tokenizer.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/special_tokens_map.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/recipe.yaml with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/preprocessor_config.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/model.safetensors.index.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/model-00002-of-00002.safetensors with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/model-00001-of-00002.safetensors with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/merges.txt with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/generation_config.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/config.json with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/chat_template.jinja with huggingface_hub

Delete (Nanonets-OCR2-3B_3epochs_lr1e-05_(fft)_batch32-prompt_generic-lr1e-05_3epochs_dpo)->(train)|awq(FP8, samples=1000)/added_tokens.json with huggingface_hub

Upload folder using huggingface_hub

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Upload 2 files

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Delete images/logo_horizontal_black_whiteback.png

Delete images/logo_dharma_ai.png