Meituan releases a 1.6-trillion-parameter agent model that never touched an Nvidia GPU
The Decoder ·
Summary
Meituan has released the weights for LongCat-2.5, a 1.6-trillion-parameter model built for AI agents and trained entirely on Chinese hardware. On Terminal Bench, it scores 88.4 points, landing just behind Claude Opus 5. The sample deployment runs on 194 Huawei Ascend chips, and M
- Source
- The Decoder
- Published (IST)
Original reporting belongs to The Decoder. DEEKOYA links the headline and a short excerpt only; we do not republish the full article.
Explore more: Today in AI · Research papers · AI tools · Learn AI
Related stories
- GlobeNewswire
A2E AI Launches AI Video Generator With All-in-One Video and Image Models
GlobeNewswireTools
- Business Insider
What job apocalypse? AI could add almost 2 million new jobs on net by 2030.
Business InsiderBusiness
- Foro3D
Autonomous AI agents fail OpenAI cybersecurity audit
Foro3DAI
- www.tokenpost.com
Tencent Adds Gemini 4 Carbon to WorkBuddy for Pro Plan Users
www.tokenpost.comModels
Unite.AI
Megaspeed Retains Counsel to Probe Reports on Its AI Cloud Business
CNBC reported on November 15, 2026, that Megaspeed had retained external legal counsel to investigate media reports…
Unite.AIBusiness
- The 420.in
South Korea’s Top Tax Exam Scorer Admits Cheating Using Generative AI
The 420.inAI
Who reported “Meituan releases a 1.6-trillion-parameter agent model that never touched an…”?
The Decoder, on 15 Nov 2026. The full article is at the source link on this page.
Comments
No comments yet. Share your own experience below.
Visitor posts are moderated before they appear and are the opinion of their authors, not DEEKOYA AI. Names are not verified. See our privacy policy.