💻 Technology 12h ago · TechNode Feed

WeChat AI Team Details WeLM Models Scaling to 617B Parameters

TechNode
TechNode China tech
View Channel →
WeChat AI Team Details WeLM Models Scaling to 617B Parameters
Source ↗ 👁 0 💬 0
Tencent’s WeChat AI team has detailed a new scaling approach for its WeLM model family. The team trained WeLM-HD4-80B and WeLM-HD4-617B models using a method called Hidden Decoding, which expands each token into multiple internal computation streams without increasing the main Transformer backbone.
The 80B model activates 3 billion parameters, while the 617B model activates 23 billion. Both models outperformed their matched autoregressive baselines across nine shared benchmarks in the team’s tes

Comments (0)

Sign in to join the discussion

More Like This

I replaced Canva, Adobe, and PowerPoint with a tool that's been hiding in plain sight, and it does everything they charge for
XDA · 2h ago
Tecno Pova 8 Pro unveiled with an Alive Matrix Display, P1 Graphic Chip, and 144FPS gaming support
GSMArena.com - Latest articles · 2h ago
Apple TV gets big batch of classic movies you can stream at no extra cost
Cult of Mac · 3h ago
[Update] Shiprocket IPO Oversubscribed 3.16X On Day 2
Inc42 Media · 3h ago
SK Hynix is spending $720bn as its shares fall by half
The Next Web · 3h ago
iOS 27 gives Apple Photos two features we’ve waited a long time for
9to5Mac · 3h ago