OpenAI Jalapeño: First Custom AI Inference Chip
June 27, 2026
OpenAI unveiled Jalapeño, its first custom AI chip — an LLM inference ASIC co-designed with Broadcom in nine months to reduce its reliance on Nvidia GPUs.
#ai-hardware
Huawei Ascend 950PR Beats NVIDIA H20: 2.8× FP8, CUDA-Ready
April 2, 2026
Huawei Ascend 950PR delivers 1 PFLOPS FP8 — 2.8× NVIDIA H20 — and runs CUDA via CANN Next. ByteDance + Alibaba ordered; 750K units shipping in 2026.
The Custom AI Chip Race 2026: Big Tech vs Nvidia
March 25, 2026
The custom AI chip race in 2026 — Meta MTIA, Google TPU, Amazon Trainium, Microsoft Maia vs. Nvidia. Strategy, cost, supply, and workload fit for each.
Neuromorphic Computing: Brain-Inspired AI
February 25, 2026
Explore how neuromorphic computing is redefining artificial intelligence with brain-inspired chips, spiking neural networks, and real-world use cases.