Guoguang Quantum Unveils Xuanmi: Industry’s First Quantum-Enhanced Large Model Family

On August 29, Guoguang Quantum announced via its official WeChat account that its Super Artificial Intelligence research team has integrated quantum technologies into the decision-making and reasoning processes of large language models, launching the Xenomi (Xenomi) Large Model Family tailored for scientific research and academic applications. Compared to mainstream open-source models, this model family demonstrates more stable domain-specific comprehension, clearer task judgment, and shorter decision pathways in tasks such as science communication, routing, and agent-based decision-making. Guoguang Quantum employed a domain-specific model training approach, independently constructing massive datasets and conducting extensive fine-tuning and inference testing based on state-of-the-art (SOTA) open-source foundation models. Through repeated validation in real-world research scenarios, the company has developed the Xenomi large model family optimized for diverse tasks. In applications like knowledge retrieval and content review, the model family delivers more precise domain-specific judgments within predefined task boundaries. Currently, the Xenomi large models have been deployed in Guoguang Quantum’s Quantum Computing Intelligent Experiment Platform and the Quamido Agent System, where they accurately respond to specific commands in modern AI software and rapidly generate decision outputs and review results. Within the Quamido Agent System, Xenomi serves as the foundational domain model, supporting key workflows including literature search, experimental planning, hybrid programming, quantum task scheduling, data analysis, and report generation.