Instructions to use sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("lightonai/LightOnOCR-2-1B")
model = PeftModel.from_pretrained(base_model, "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2")

Transformers

How to use sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2 with Transformers:

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

pipe = pipeline("text-generation", model="sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2

SGLang

How to use sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2 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 "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2 with Docker Model Runner:
```
docker model run hf.co/sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2
```

LightOnOCR-2-1B-Math-Handwritten-v2

This model is a fine-tuned version of lightonai/LightOnOCR-2-1B on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2029

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 6e-05
train_batch_size: 4
eval_batch_size: 6
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 10
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.1245	0.0348	50	0.5313
0.3920	0.0696	100	0.3782
0.2918	0.1044	150	0.3329
0.2947	0.1392	200	0.3074
0.2567	0.1740	250	0.2927
0.2456	0.2088	300	0.2744
0.2545	0.2436	350	0.2635
0.2225	0.2784	400	0.2543
0.2311	0.3132	450	0.2482
0.2350	0.3480	500	0.2388
0.2025	0.3828	550	0.2412
0.2146	0.4176	600	0.2342
0.1878	0.4524	650	0.2332
0.1867	0.4872	700	0.2276
0.2013	0.5220	750	0.2237
0.1939	0.5568	800	0.2208
0.1795	0.5916	850	0.2205
0.1977	0.6264	900	0.2150
0.1969	0.6612	950	0.2120
0.1780	0.6960	1000	0.2126
0.1747	0.7308	1050	0.2114
0.1736	0.7656	1100	0.2105
0.1885	0.8004	1150	0.2075
0.1792	0.8352	1200	0.2063
0.1947	0.8700	1250	0.2051
0.1738	0.9048	1300	0.2043
0.1805	0.9396	1350	0.2035
0.1608	0.9744	1400	0.2029

Framework versions

PEFT 0.18.1
Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.8.4
Tokenizers 0.22.2

Downloads last month: 8

Model tree for sprok-daniel-mozaik/LightOnOCR-2-1B-Math-Handwritten-v2

Base model

lightonai/LightOnOCR-2-1B

Adapter

(8)

this model