MiniMax M2.5 Launch Rivals Top AI Models, Valuation Exceeds $180B

Another major breakthrough has emerged in the domestic AI landscape. On February 12, 2026, MiniMax officially launched its latest flagship programming large model, MiniMax M2.5. Positioned as "the world's first production-level model natively designed for Agent scenarios," its release not only stirred the tech community but also ignited significant activity in the capital markets.
Peak Performance: A Direct Competitor to Claude Opus 4.6
MiniMax M2.5 demonstrates formidable strength in programming and agent capabilities (Coding & Agentic):
Full-Stack Programming: It supports end-to-end development for PC, App, and cross-platform applications, showcasing deep programming expertise.
The Productivity Powerhouse: In core office scenarios like advanced Excel processing, in-depth research, and PPT creation, it operates at a state-of-the-art (SOTA) level.
Global Benchmarking: Officially, the model's overall performance is stated to directly compete with top-tier international models such as Claude Opus 4.6.
The Efficiency Advantage: Compact Size, High Throughput
Diverging from the trend of simply scaling parameters, the M2.5 model adopts a highly efficient approach:
Streamlined Architecture: It utilizes only 10B active parameters, dramatically reducing memory footprint.
High Performance per Watt: It achieves an impressive throughput of 100 TPS, with inference speed rivaling even some leading global models.
Market Frenzy: Valuation Soars Past HKD 180 Billion
Bolstered by the positive news of the new model launch, MiniMax's share price surged over 20% in a single day, pushing its total market capitalization above HKD 180 billion. This stellar performance further cements its strong standing in the competition for the title of premier AI stock against rivals like Zhipu.
The launch of MiniMax M2.5 signals a shift for domestic large models from "catching up broadly" to "reshaping specialized scenarios," heralding a new chapter for AI-powered programming and Agent applications.
Related article
Cyberspace Administration of China mandates tagging of AI-generated and fictional short videos
The Cyberspace Administration of China has rolled out a comprehensive plan to standardize short video content labeling, mandating that platforms offer six required tags—including "AI-generated content"—ushering in a new era of mandatory transparency
DeepL, renowned for text translation, now targets voice translation
DeepL, a translation company best known for its text-based tools, has launched a voice-to-voice translation suite today that addresses scenarios such as meetings, mobile and web conversations, and group discussions for frontline workers through custo
Talat’s AI meeting notes live on your device, not the cloud
Granola, the AI-powered notetaking app valued at $250 million, has gained traction among tech founders and venture capitalists. But one developer sees demand for a more private, fully local alternative available for a one-time fee with no subscriptio
Related Special Topic Recommendations
Comments (0)
0/500

Another major breakthrough has emerged in the domestic AI landscape. On February 12, 2026, MiniMax officially launched its latest flagship programming large model, MiniMax M2.5. Positioned as "the world's first production-level model natively designed for Agent scenarios," its release not only stirred the tech community but also ignited significant activity in the capital markets.
Peak Performance: A Direct Competitor to Claude Opus 4.6
MiniMax M2.5 demonstrates formidable strength in programming and agent capabilities (Coding & Agentic):
Full-Stack Programming: It supports end-to-end development for PC, App, and cross-platform applications, showcasing deep programming expertise.
The Productivity Powerhouse: In core office scenarios like advanced Excel processing, in-depth research, and PPT creation, it operates at a state-of-the-art (SOTA) level.
Global Benchmarking: Officially, the model's overall performance is stated to directly compete with top-tier international models such as Claude Opus 4.6.
The Efficiency Advantage: Compact Size, High Throughput
Diverging from the trend of simply scaling parameters, the M2.5 model adopts a highly efficient approach:
Streamlined Architecture: It utilizes only 10B active parameters, dramatically reducing memory footprint.
High Performance per Watt: It achieves an impressive throughput of 100 TPS, with inference speed rivaling even some leading global models.
Market Frenzy: Valuation Soars Past HKD 180 Billion
Bolstered by the positive news of the new model launch, MiniMax's share price surged over 20% in a single day, pushing its total market capitalization above HKD 180 billion. This stellar performance further cements its strong standing in the competition for the title of premier AI stock against rivals like Zhipu.
The launch of MiniMax M2.5 signals a shift for domestic large models from "catching up broadly" to "reshaping specialized scenarios," heralding a new chapter for AI-powered programming and Agent applications.
Cyberspace Administration of China mandates tagging of AI-generated and fictional short videos
The Cyberspace Administration of China has rolled out a comprehensive plan to standardize short video content labeling, mandating that platforms offer six required tags—including "AI-generated content"—ushering in a new era of mandatory transparency
DeepL, renowned for text translation, now targets voice translation
DeepL, a translation company best known for its text-based tools, has launched a voice-to-voice translation suite today that addresses scenarios such as meetings, mobile and web conversations, and group discussions for frontline workers through custo
Talat’s AI meeting notes live on your device, not the cloud
Granola, the AI-powered notetaking app valued at $250 million, has gained traction among tech founders and venture capitalists. But one developer sees demand for a more private, fully local alternative available for a one-time fee with no subscriptio





Home






