Chinese AI startup MiniMax plans to open-source a 2.7 trillion parameter model later this year

2026-07-08 · Source: The Decoder · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Fundamental Awareness, quick

Summary

Chinese AI startup MiniMax is reportedly planning to open-source a new large language model with 2.7 trillion parameters later this year, potentially as early as Q3. Internally named M3 Pro, this model would significantly surpass MiniMax's current 428 billion parameter M3 model and become the largest open-source AI model from China. Larger models are generally expected to enhance performance on complex reasoning and multi-step instruction tasks. This development comes as Chinese open-source models gain traction among developers seeking cost-effective solutions for high-volume, less critical tasks, positioning MiniMax against competitors like Zhipu, DeepSeek, and Moonshot AI. However, the release timing could be influenced by recent reports indicating the Chinese government's intent to tighten controls on future AI model exports.

Key takeaway

For AI Directors and ML Engineers evaluating large language models, MiniMax's planned 2.7 trillion parameter open-source model, M3 Pro, presents a significant new option. You should monitor its Q3 release for potential cost-effective deployment in complex reasoning or high-volume, less critical applications. Be aware that evolving Chinese government export controls might impact its availability or future use.

Key insights

MiniMax plans to open-source a 2.7 trillion parameter model, potentially shifting the landscape for large Chinese AI models.

Principles

• Larger models often improve complex reasoning.
• Open-source models offer cost-effective solutions.

In practice

• Evaluate MiniMax's M3 Pro for high-volume tasks.
• Monitor Chinese AI export regulations.

Topics

• MiniMax
• Large Language Models
• Open-source AI
• AI Model Parameters
• Chinese AI
• AI Export Controls

Best for: CTO, VP of Engineering/Data, AI Architect, AI Scientist, Director of AI/ML, Tech Journalist

Related on AIssential

• MiniMax's new open M2.5 and M2.5 Lightning near state-of-the-art while costing 1/20th of Claude Opus 4.6 — MiniMax's M2.5 models offer near state-of-the-art AI performance at a fraction of the cost, enabling widespread agentic applications.
• MiniMax M3 IS INSANE! BEST Opensource AI Model! Beats Opus 4.7 and 50x Cheaper! (Fully Tested) — MiniMax M3 is an open-weight, natively multimodal model offering frontier performance in coding and agentic tasks at significantly lower costs.
• How are Chinese models so strong with so little investment? — Chinese AI models achieve high performance through distillation from Western models and a robust national AI ecosystem.
• MiniMax M2.7 matches GPT-5.3-Codex in software engineering tasks — Chinese AI labs are increasingly releasing large language model weights, though licensing terms vary significantly.
• ChinAI #349: Tokens Made in China? — Chinese AI models are gaining traction among U.S. developers due to their compelling price-to-performance ratio.
• 40 KB Conversational AI Model — Extremely compact AI models and advanced agentic capabilities are expanding the frontiers of AI deployment and development.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Decoder.