Category
NVIDIA has released Nemotron 3.5 Lightning, which activates 3B out of a 30B total parameter count. The design hands planning to large models while routing repetitive execution to a fast, small model.
Riot will provide 191MW of AI data center capacity in Rockdale over 20 years. Bloomberg reports the tenant is Anthropic, with total base rent reaching $9.1 billion.
OpenAI has released a preview of the Linux version of its desktop app that integrates ChatGPT, Work, and Codex. It covers five releases including Ubuntu, Debian, and Fedora, but feature parity and distribution methods remain issues to confirm going forward.
A new notation called "Indirect Einsum," which embeds the positions of zeros in sparse arrays directly into expressions, has enabled a team from MIT CSAIL and Georgia Tech to speed up sparse computation on GPUs by up to 3.81x while cutting code volume by up to 4,491x. By reusing existing dense matrix compilers, the design opens a path for developers without sparse-computation expertise.
Microsoft's image generation model MAI-Image-2.6 has risen to 2nd place in Arena's text-to-image generation ranking. We examine the improvements across all 8 categories and the remaining gaps in pricing and specifications before product rollout.
Anthropic will embed invisible watermarks into the generated text of new Claude models offered in the EU starting August 2, extending the feature to APIs and Claude Code worldwide. Existing models and short-text detection remain in transition and under validation.
A hardware/software co-design accepted to MICRO 2026 uses the signal of "which task is currently being executed" in a multi-task inference model to physically skip unnecessary compute tiles on an FPGA. It cuts FLOPs by up to 76% and shortens latency by 2.4x, while the contrast with the same mask causing a 22% slowdown on GPUs clearly reveals the hardware requirements for exploiting sparsity.
NVIDIA has signed memoranda of understanding with six financial firms, outlining a plan to mobilize over $500 billion in third-party capital for AI compute infrastructure. While GPUs are treated as financial assets, the definitive agreements and loss-sharing arrangements remain undetermined.
Intel has proposed a new $15 billion stock offering, but the use of proceeds is for general purposes, and no allocation to 14A has been disclosed. While 14A is proceeding toward risk production for Intel's own products in second-half 2027 and a 2028 ramp, no external customer contracts have been confirmed.
Reports indicate Windows OEM license fees rose by an average of 7-10% starting July 2026. With memory costs already high, this adds to software expenses, and Taiwan is expected to see PC prices rise roughly 5% in Q3.
