Companies like Meta, NVIDIA, and AMD are racing to design proprietary AI chips—from inference accelerators to novel CPU cores—to reduce reliance on standard GPUs and gain control over performance and costs.
DeepSeek is reportedly developing its own AI processor, a strategic move toward technological self-reliance and greater control over its inference pipeline, sources indicate.
NVIDIA has introduced initial support for the new Arm "Rigel" core in both GCC and LLVM Clang, ahead of the Rosa CPU launch. The move accelerates its vertical integration strategy for AI workloads, with potential impact on efficiency and sovereignty ...
Meta will begin manufacturing its latest in-house AI chips in September, aiming to reduce reliance on costly Nvidia GPUs. The move underscores a structural shift toward vertical integration in AI hardware, with significant implications for the infere...
Meta will start producing its first AI chip from the MTIA line in September, with the goal of doubling computing capacity across its data centers. The move reduces dependence on NVIDIA and marks a turning point for custom hardware in large-scale AI.
Nvidia partners with startup d-Matrix, a maker of AI inference chips, on a joint system that combines GPUs with dedicated accelerators. AI cloud firm Parasail is the first customer. The move signals a strategy to dominate inference without crushing r...
Samsung’s first PCIe 6.0 eSSDs for the Nvidia Vera Rubin platform signal a hierarchy shift: storage is no longer a secondary bottleneck but an active memory component for on-premise inference. The move reveals that the next link in the AI chain is th...
While touting the single-thread performance of its Vera CPU with Olympus cores, NVIDIA confirmed some details about the upcoming Rosa and its Rigel core. The company takes a decisive step toward a proprietary CPU, with profound implications for those...
Turing Inc. has secured investment from AMD Ventures and started using AMD chips for its autonomous vehicle AI, gradually moving away from Nvidia. The move signals crucial hardware diversification for edge deployment, where cost, latency, and direct ...
