What is SmartResume
SmartResume is an intelligent resume parsing tool open-sourced by Alibaba. Through OCR text extraction, PDF metadata parsing, layout detection, and LLM intelligent structured processing, SmartResume automatically converts resumes in formats such as PDF, images, and Word into structured data, such as name, education, and work experience. The tool uses YOLOv10 for layout detection and a fine-tuned Qwen3-0.6B for content structuring. It supports API and local model deployment and can complete parsing in seconds. It is widely used in enterprise recruitment, recruitment platforms, campus recruitment, and headhunting agencies, greatly improving HR work efficiency.
SmartResume's Main Features
-
Multi-format resume parsing: Supports parsing resumes in common formats such as PDF, images, Word, and Excel.
-
OCR and PDF metadata extraction: Combines OCR technology and PDF metadata parsing to extract plain text content from documents.
-
Layout detection and reading order reconstruction: Uses the YOLOv10 model to detect resume layout, identify regions such as titles, paragraphs, and tables, and reconstruct the logical reading order.
-
LLM intelligent structured processing: Based on a fine-tuned Qwen3-0.6B model, converts text content into structured JSON format and extracts key fields (such as name, phone, email, work experience, etc.).
-
Local model deployment: Supports local model deployment, reducing dependence on external APIs and ensuring data privacy.
-
API invocation and integration: Provides a Python API and command-line interface, making it easy to integrate with recruitment systems and achieve automated resume processing.
How to Use SmartResume
-
Environment preparation: Ensure the device meets the minimum requirements, such as Python 3.9 or above, 8GB or more memory, and 10GB or more storage space (optional CUDA 11.0 or above for GPU acceleration).
-
Clone the repository: Use Git to clone the SmartResume project https://github.com/alibaba/SmartResume克隆到本地。
-
Create an environment: Use Conda to create a Python environment and activate it.
-
Install dependencies: Run commands to install the dependency packages required by the project.
-
Configuration file: Copy the configuration file template and modify it as needed, adding necessary information such as API keys.
-
Start parsing: Call the parsing function through the command-line interface (CLI) or Python API, specifying the resume file path and the field types to extract.
-
Local model deployment(optional): If local model deployment is needed, download the model files and start the local model service to reduce dependence on external APIs.
SmartResume's Project Address
- GitHub repository：https://hf-mirror.com/Alibaba-EI/SmartResume
SmartResume's Application Scenarios
-
Enterprise recruitment systems: Automatically parse resumes submitted by candidates, extract key information, and directly populate it into the enterprise's human resource management system.
-
Recruitment platforms: Quickly tag and screen massive resumes, helping recruiters quickly find candidates who meet job requirements.
-
Campus recruitment: Supports batch import of student resumes, efficiently matches job requirements, and screens out qualified candidates.
-
Headhunting agencies: Enables structured management of candidate data, achieving precise matching and recommendations, and improving service quality.
-
HR SaaS products: SmartResume provides intelligent resume entry functionality and supports API calls, making it easy to integrate into HR SaaS products.