Downloads · 30 days
27
12% of all-time downloads
AlexandreSheva/rukopys-qwen3-vl-8b-page-a100-v2
rukopys-qwen3-vl-8b-page-a100-v2 is a text generation model from AlexandreSheva. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
Improved page-level LoRA adapter for Qwen/Qwen3-VL-8B-Instruct, fine-tuned for Ukrainian handwritten document parsing.
Downloads · 30 days
27
12% of all-time downloads
All-time downloads
226
Public
Repo size
2.1 GB
Likes
1
Public
Click a slice to open those files.
.safetensors349 MB · 97%
From the Hugging Face model README
Improved page-level LoRA adapter for Qwen/Qwen3-VL-8B-Instruct, fine-tuned for Ukrainian handwritten document parsing.
This is the preferred public checkpoint from the RUKOPYS HTR project. It improves on the original page-level adapter with an A100-class training run focused on full-page image-to-JSON extraction.
The model reads a scanned handwritten page and predicts structured document regions with bounding boxes, region types, language metadata, and transcribed text.
Most OCR pipelines treat handwriting recognition as plain text extraction. This adapter is trained for a more practical document AI task: recovering both layout and text from Ukrainian handwritten pages in one structured output.
It is part of a reproducible HTR pipeline covering:
Use this checkpoint instead of the original adapter when comparing the two public RUKOPYS Qwen3-VL 8B LoRA releases.
Original baseline:
AlexandreSheva/rukopys-qwen3-vl-8b-page
Trained on:
AlexandreSheva/rukopys-curated-mvp
The curated dataset converts the original RUKOPYS release into task-ready artifacts for page-level VLM fine-tuning, crop-level transcription, and layout detection.
The expected output is structured JSON:
[
{
"bbox": [10, 20, 300, 80],
"type": "handwritten",
"language": "uk",
"text": "..."
}
]
Focused on Ukrainian handwritten documents. Performance depends on scan quality, page resolution, and handwriting style. Bounding boxes and transcription should be evaluated together. The adapter is Apache-2.0, but the training data derives from a CC BY-NC-SA 4.0 dataset.
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4169 | 0.3342 | 100 | 0.4485 |
| 0.4174 | 0.6683 | 200 | 0.4333 |
| 0.3848 | 1.0 | 300 | 0.4249 |
| 0.3604 | 1.3342 | 400 | 0.4200 |
| 0.3535 | 1.6683 | 500 | 0.4170 |
| 0.3476 | 2.0 | 600 | 0.4150 |
| 0.3405 | 2.3342 | 700 | 0.4202 |
| 0.3261 | 2.6683 | 800 | 0.4186 |
| 0.3139 | 3.0 | 900 | 0.4243 |
| 0.3208 | 3.3342 | 1000 | 0.4218 |
| 0.3002 | 3.6683 | 1100 | 0.4251 |
| 0.3070 | 4.0 | 1200 | 0.4232 |