Instructions to use TomasFAV/DonutInvoiceCzechV0123 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TomasFAV/DonutInvoiceCzechV0123 with Transformers:

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

pipe = pipeline("image-text-to-text", model="TomasFAV/DonutInvoiceCzechV0123")

# Load model directly
from transformers import AutoTokenizer, AutoModelForImageTextToText

tokenizer = AutoTokenizer.from_pretrained("TomasFAV/DonutInvoiceCzechV0123")
model = AutoModelForImageTextToText.from_pretrained("TomasFAV/DonutInvoiceCzechV0123")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use TomasFAV/DonutInvoiceCzechV0123 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TomasFAV/DonutInvoiceCzechV0123"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TomasFAV/DonutInvoiceCzechV0123",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TomasFAV/DonutInvoiceCzechV0123

SGLang

How to use TomasFAV/DonutInvoiceCzechV0123 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 "TomasFAV/DonutInvoiceCzechV0123" \
    --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": "TomasFAV/DonutInvoiceCzechV0123",
		"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 "TomasFAV/DonutInvoiceCzechV0123" \
        --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": "TomasFAV/DonutInvoiceCzechV0123",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TomasFAV/DonutInvoiceCzechV0123 with Docker Model Runner:
```
docker model run hf.co/TomasFAV/DonutInvoiceCzechV0123
```

DonutInvoiceCzechV0123

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1672
Accuracy: 0.9460
F1: 0.9214

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: 9e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
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
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.2086	1.0	46	0.1679	0.9141	0.8816
0.1011	2.0	92	0.1561	0.9371	0.9128
0.0632	3.0	138	0.1630	0.9386	0.9111
0.0376	4.0	184	0.1921	0.8951	0.8927
0.0286	5.0	230	0.1798	0.9291	0.9094
0.0214	6.0	276	0.1672	0.9460	0.9214
0.0159	7.0	322	0.1734	0.9432	0.9214
0.0223	8.0	368	0.1815	0.9352	0.9074
0.0071	9.0	414	0.1857	0.9304	0.9012
0.0175	10.0	460	0.1976	0.9288	0.9088
0.0073	11.0	506	0.1983	0.9439	0.9111
0.0034	12.0	552	0.2064	0.9318	0.9090
0.0077	13.0	598	0.2131	0.9359	0.9072
0.0050	14.0	644	0.2108	0.9369	0.9072
0.0007	15.0	690	0.2074	0.9340	0.9072
0.0006	16.0	736	0.2084	0.9357	0.9106
0.0011	17.0	782	0.2075	0.9363	0.9128
0.0008	18.0	828	0.2030	0.9384	0.9192
0.0009	19.0	874	0.2025	0.9353	0.9192
0.0006	20.0	920	0.2027	0.9338	0.9175

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.0.0
Tokenizers 0.22.2

Downloads last month: 91

Safetensors

Model size

0.2B params

Tensor type

I64

F32

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support