> ## Documentation Index > Fetch the complete documentation index at: https://docs.lapathoniia.top/llms.txt > Use this file to discover all available pages before exploring further. # Rukopys OCR > Ukrainian handwritten document recognition ## What is Rukopys? Rukopys is a specialised model for recognising **Ukrainian handwritten text**. Supports images and PDF documents. **Supported formats:** PNG, JPG, JPEG, WebP, PDF, TIF, TIFF (up to 50 MB) **Quota:** 4 images/hour (default). Quota applies **per page** for multi-page files. ## Via Chat UI Go to [app.lapathoniia.top](https://app.lapathoniia.top), select **Rukopys OCR** model, attach an image or PDF. ## Via API ```python theme={null} import httpx, base64, json with open("handwritten.jpg", "rb") as f: image_b64 = base64.b64encode(f.read()).decode() resp = httpx.post( "https://app.lapathoniia.top/chat/stream", headers={"Authorization": "Bearer YOUR_API_KEY"}, json={ "message": "Recognise text", "model": "rukopys-ocr", "session_id": "ocr-1", "user_id": "user-123", "web_search": False, "connectors": False, "chat_history": [], "images": [{"base64": image_b64, "mime_type": "image/jpeg"}], }, ) result = [] for line in resp.iter_lines(): if line.startswith("data: "): d = json.loads(line[6:]) if "text" in d: result.append(d["text"]) regions = json.loads("".join(result)) ``` ### Response Format ```json theme={null} [ { "text": "Dear Sir,", "bbox": [120, 45, 380, 70], "confidence": 0.94, "page": 1 }, { "text": "We are forwarding the documents", "bbox": [120, 80, 520, 105], "confidence": 0.91, "page": 1 } ] ``` | Field | Description | | - | - | | `text` | Recognised text of the region | | `bbox` | Coordinates \[x1, y1, x2, y2] in pixels | | `confidence` | Model confidence (0–1) | | `page` | Page number (for PDFs) | ## OCR + LLM Pipeline Combine Rukopys OCR with a language model — OCR via `/chat/stream`, then extraction with MamayLM. ```python theme={null} import httpx, base64, json def ocr_then_extract(image_path: str, prompt: str, api_key: str) -> dict: with open(image_path, "rb") as f: b64 = base64.b64encode(f.read()).decode() headers = {"Authorization": f"Bearer {api_key}"} # Step 1 — OCR ocr_resp = httpx.post( "https://app.lapathoniia.top/chat/stream", headers=headers, json={ "message": "Розпізнай текст", "model": "rukopys-ocr", "session_id": "ocr-pipeline", "user_id": "agent", "web_search": False, "connectors": False, "chat_history": [], "images": [{"base64": b64, "mime_type": "image/jpeg"}], }, ) ocr_text = "".join( json.loads(l[6:])["text"] for l in ocr_resp.iter_lines() if l.startswith("data: ") and "text" in json.loads(l[6:]) ) # Step 2 — Extract structured data llm_resp = httpx.post( "https://app.lapathoniia.top/chat/stream", headers=headers, json={ "message": f"{prompt}\n\nText:\n{ocr_text}", "model": "MamayLM-Gemma-3-27B-IT-v2.0", "session_id": "ocr-pipeline", "user_id": "agent", "web_search": False, "connectors": False, "chat_history": [], }, ) result_text = "".join( json.loads(l[6:])["text"] for l in llm_resp.iter_lines() if l.startswith("data: ") and "text" in json.loads(l[6:]) ) return json.loads(result_text) # Example: grade extraction from handwritten tests result = ocr_then_extract( "test.jpg", "Extract: student name, date, subject, score (0-100). Return JSON only.", api_key="sk-...", ) # → {"student": "Ivan Petrenko", "date": "2024-05-12", "subject": "Math", "score": 87} ``` Other pipeline use cases: * `"Extract: date, signatories, amounts. Return JSON only."` — contracts * `"Summarize this handwritten note in 2 sentences."` — meeting notes * `"Translate to English."` — Ukrainian handwritten → English ## Limits * **4 images/hour** — default quota per key * **50 MB** — maximum file size * **50 pages** — maximum pages per PDF/TIFF * Check current quota: [`GET /user/limits`](/api/usage-limits) ## What is Rukopys? Rukopys — спеціалізована модель для розпізнавання **українських рукописних текстів**. Підтримує зображення та PDF. **Формати:** PNG, JPG, JPEG, WebP, PDF, TIF, TIFF (до 50 МБ) **Квота:** 4 зображення/год. Квота рахується **по сторінках** для багатосторінкових файлів. ## Via API ```python theme={null} import httpx, base64, json with open("manuscript.jpg", "rb") as f: b64 = base64.b64encode(f.read()).decode() resp = httpx.post( "https://app.lapathoniia.top/chat/stream", headers={"Authorization": "Bearer YOUR_API_KEY"}, json={ "message": "Розпізнай текст", "model": "rukopys-ocr", "session_id": "ocr-1", "user_id": "user-123", "web_search": False, "connectors": False, "chat_history": [], "images": [{"base64": b64, "mime_type": "image/jpeg"}], }, ) ``` ### Response Format ```json theme={null} [ { "text": "Шановний пане,", "bbox": [120, 45, 380, 70], "confidence": 0.94, "page": 1 } ] ``` ## OCR + LLM Pipeline Поєднайте Rukopys OCR з мовною моделлю — OCR через `/chat/stream`, потім витяг через MamayLM. ```python theme={null} import httpx, base64, json def ocr_та_витяг(шлях: str, інструкція: str, api_key: str) -> dict: with open(шлях, "rb") as f: b64 = base64.b64encode(f.read()).decode() headers = {"Authorization": f"Bearer {api_key}"} # Крок 1 — OCR ocr_resp = httpx.post( "https://app.lapathoniia.top/chat/stream", headers=headers, json={ "message": "Розпізнай текст", "model": "rukopys-ocr", "session_id": "ocr-pipeline", "user_id": "agent", "web_search": False, "connectors": False, "chat_history": [], "images": [{"base64": b64, "mime_type": "image/jpeg"}], }, ) ocr_text = "".join( json.loads(l[6:])["text"] for l in ocr_resp.iter_lines() if l.startswith("data: ") and "text" in json.loads(l[6:]) ) # Крок 2 — Витяг структурованих даних llm_resp = httpx.post( "https://app.lapathoniia.top/chat/stream", headers=headers, json={ "message": f"{інструкція} Текст: {ocr_text}", "model": "MamayLM-Gemma-3-27B-IT-v2.0", "session_id": "ocr-pipeline", "user_id": "agent", "web_search": False, "connectors": False, "chat_history": [], }, ) result_text = "".join( json.loads(l[6:])["text"] for l in llm_resp.iter_lines() if l.startswith("data: ") and "text" in json.loads(l[6:]) ) return json.loads(result_text) # Приклад: витяг оцінок з рукописних контрольних result = ocr_та_витяг( "test.jpg", "Витягни: ім’я учня, дату, предмет, оцінку (0-100). Поверни тільки JSON.", api_key="sk-...", ) # → {"student": "Іван Петренко", "date": "2024-05-12", "subject": "Математика", "score": 87} ``` Інші варіанти використання: * `"Витягни: дату, підписантів, суми. JSON."` — договори * `"Стисло переказ у 2 реченнях."` — нотатки з нарад * `"Перекладіть англійською."` — рукопис → англійська