Qwen News & Updates

Mistral releases a family of 10 open-source AI models, including Mistral Large 3 and Ministral 3, designed to run on laptops, drones, and edge devices, prioritizing flexibility, efficiency, and distributed intelligence.

Arcee releases Trinity Mini and Trinity Nano Preview, two open-source AI models under Apache 2.0 license, aiming to regain U.S. control in open-source AI development.

Liquid AI releases a 51-page technical report detailing the architecture and training data behind its efficient on-device AI models, making them open-source for other organizations to use.

Notable AI releases and advancements in 2025 included OpenAI's continued growth, China's open-source wave, and growth of local models.

Nvidia and the University of Hong Kong release Orchestrator, an 8-billion-parameter model that coordinates tools and LLMs for complex problem-solving. This model outperforms larger models at a lower cost and improves efficiency.

Amazon Bedrock has expanded its list of serverless models, adding 18 new open-weight models from various AI companies, including the new Mistral Large 3 and Ministral 3 models.

OpenMMReasoner, a new training framework, boosts AI multimodal reasoning capabilities while using smaller and smarter datasets. It provides a step-by-step process for reproducibility and improves transparency in the training pipeline.

University of Science and Technology researchers developed Agent-R1, a new reinforcement learning (RL) framework for large language models (LLMs) that addresses challenges in complex agentic tasks.

Tech companies are releasing AI-powered glasses for consumers, offering various features such as search, payments, and language translation.