Aug 5 — Ant Group’s Bailing open-sourced Ling-3.0-flash, a next-generation
native hybrid inference model designed for high model density, full-chain agent
closed-loop operation and low-cost scalable deployment. Huawei’s Ascend
completed zero-day adaptation and launched the CANN PyPTO operator programming
framework to shorten fused-operator delivery cycles, accelerate operator
integration and inference-framework performance tuning, and published a full
deployment engineering stack for developers, government and enterprise customers
to bolster China’s domestic large-model ecosystem.