Scientific and Technical Research Improving Search Query Rewriting Quality with Small Language Models (SLMs) and Recurrent Reasoning Developing a novel architecture at the intersection of information retrieval and language model...
Iran's First Specialized Center for Large Language Models (LLMs)
The first specialized center in the field of large language models in Iran
Mabna is the result of a collaborative effort by Iranian AI researchers, specialists, and developers who work together with a strong foundation of knowledge, experience, and a research-driven mindset. We believe that progress in artificial intelligence is not achieved through technology alone; it also requires creative and curious people who can transform knowledge into practical, real-world solutions.
Our primary focus is on training, optimizing, and customizing foundational AI models for a wide range of applications and projects. By leveraging modern technologies, we develop intelligent solutions tailored to the unique needs of businesses and organizations. Our expertise spans data processing, automation, information analysis, and the development of intelligent systems that drive efficiency, innovation, and growth.
An AI-powered assistant that operates like a trained expert, providing accurate and reliable answers to employees, managers, and customers around the clock. It retrieves information directly from your organization's documents and knowledge sources to deliver trustworthy, context-aware responses.
Our AI agents go beyond answering questions. They can perform real organizational tasks by integrating with different software systems and executing multi-step workflows automatically, acting as efficient digital employees.
We customize and fine-tune AI models using your organization's specialized knowledge, terminology, and documentation, enabling them to function as experienced domain experts rather than generic AI assistants.
Documents, regulations, contracts, reports, and expert knowledge are consolidated into an intelligent knowledge base, ensuring that valuable organizational information remains accessible, searchable, and preserved even when employees leave.
The “Mobayyen” artificial intelligence system is a specialized chatbot and intelligent research assistant designed to facilitate access for researchers, legal experts, and interested users to important parliamentary documents
The “Mobayyen” artificial intelligence system is a specialized chatbot and intelligent research assistant designed to facilitate access for researchers, legal experts, and interested users to important parliamentary documents
Scientific and Technical Research Improving Search Query Rewriting Quality with Small Language Models (SLMs) and Recurrent Reasoning Developing a novel architecture at the intersection of information retrieval and language model...
Research and Development AI and Persuasion; A Step Towards Smarter and More Effective Interactions A step towards smarter and more effective interactions; investigating the impact mechanisms of large language models...
Scientific and Security Research Automated Penetration Testing Based on Agentic AI Development of a dynamic system based on Large Language Models (LLMs) for goal-oriented guidance and automation of network and...
Interview and Special Report A Story of a Qom Scientist’s Journey to Create Indigenous Artificial Intelligence Exploring the achievements of the University of Qom’s Smart Software Lab, from the Judicial...
Scientific and Technological Achievement Developing Artificial Intelligence Software in the Field of Specialized Knowledge Development of a native intelligent engine by Qom University researchers; integrating advanced natural language processing technologies...
Judiciary News Report Launch of the Judge’s Artificial Intelligence Assistant System During Wartime Actions of the Judiciary’s Statistics and Information Technology Center in developing smart judicial services, sharing data, and...
AI Technical Article The Intelligent Agent Revolution: Multi-Step Decision Making with LangGraph Exploring the transition from simple language models to Agentic Systems and the role of graph-based architecture in their...
Technical Article · Updated to June 2026 Text-to-Speech (TTS)in Today’s World From Concatenative to Flow Matching and Hybrid AR+LLM models; a comprehensive review of architectures, top 2025–2026 models, Persian challenges,...
Technical Article · Updated to June 2026 Speech-to-Text (STT / ASR)in Today’s World From HMM to Hybrid SALM Models; A Comprehensive Review of Architectures, Top 2025-2026 Models, Persian Challenges, and...
