10 results on this page
A global semiconductor-stock selloff intensified, sending Samsung and SK Hynix shares down more than 7% in Seoul. The decline followed a 5% drop in the Philadelphia Semiconductor Index as investors pulled back from high-performing chip stocks.
Waymo has published the architecture of its robotaxi onboard computer, offering its first detailed look inside the vehicle system. The company also introduced a custom 5-nanometer chip and identified seven suppliers supporting the platform.
Nvidia announced full-scale production of its Spectrum-X Ethernet Photonics co-packaged optical switch for AI factories, using 200G-per-lane links. SK hynix simultaneously published a CPO technology roadmap targeting over 100 Tb/s per node, sub-1 pJ/bit energy consumption, and sub-10 ns chip-to-chip latency.
Callosum has raised $100 million in early financing to develop software that matches specific AI tasks with suitable models and chips. Backers include the UK’s public AI fund.
A Chinese-led team from Nanjing University and the National University of Singapore developed 光语芯, an ultra-low-power intelligent vision sensor chip that directly converts light into tokens. The tokens can be fed into an encoder for image recognition.
Xiaomi stabilized in Q2 2026 while expanding its AI strategy across models, chips, and robotics. Its MiMo-V2.5 model led OpenRouter call charts, the Xuanjie chip passed mass verification, and a humanoid robot began factory operations with a reported 98% success rate.
British AI-chip startup Fractile is reportedly in advanced talks to raise funds at a $6.5 billion pre-money valuation. The proposed valuation is more than six times its level three months ago, following a chip-supply deal with Anthropic.
Marvell and Google are deepening their partnership on custom AI chip development. Marvell is giving Google the right to purchase up to $12.2 billion in Marvell stock, strengthening their long-term alignment in the AI hardware race.
Cerebras launched the CS-4, its first system to combine three wafer-scale processors in a single rack. The inference-focused machine is shipping this quarter, with Cerebras claiming up to 30 times the speed of GPU-based systems for frontier models.
China has reportedly allowed ByteDance and Tencent to import 10,000 NVIDIA H200 chips each, according to the Financial Times. Other Chinese companies could receive similar approval, potentially easing access to advanced AI computing hardware.