Tencent's New Model Claims to Surpass GLM-5.1

iamfakhrealam · x · 2026-07-07

Tencent claims its new model rivals open-source flagships 2-5 times its size. In internal blind tests, 270 experts rated it higher than GLM-5.1 on real-world tasks, with the most significant improvements seen in agent capabilities like tool calling, multi-step tasks, and web search.

Related event: Tencent Open-Sources Hunyuan Hy3 for Agentic Workloads(26 posts)→

Original post →

More from Models

Models channel →