Researchers have developed a compiler-grounded hierarchical optimization framework for Triton kernels that achieves a geometric-mean speedup of 4.35× and a median speedup of 2.73× across 37 benchmark …
DeepSeek's V4-Flash API pricing undercuts OpenAI and Anthropic, listed at $0.14 per million cache-miss input tokens and $0.28 per million output tokens, compared to OpenAI's GPT-5.6 Luna at $0.20 inpu…
Huawei plans to launch its Ascend 950 AI accelerators and Atlas 950 SuperPod systems in South Korea in the fourth quarter of 2026, with local distributor support from Hansol PNS and SK Shieldus. The m…
In April 2026, DeepSeek released its V4 model, a 1.6 trillion parameter MoE architecture, and for the first time officially validated its inference on Huawei's Ascend 950PR chip, marking a significant…