Microsoft, NVIDIA and Anthropic announce partnerships to scale Claude Artificial Intelligence
Anthropic will scale its Claude Artificial Intelligence model on Microsoft Azure using NVIDIA systems, broadening enterprise access. the companies also announced a deep technology collaboration between NVIDIA and Anthropic to optimize models and hardware.
NVIDIA and Microsoft integrate latest technologies to power artificial intelligence superfactories
NVIDIA and Microsoft expanded their collaboration at Microsoft Ignite to equip Microsoft’s Fairwater artificial intelligence superfactory with Spectrum-X switches and Blackwell GPUs. The work also includes new Azure VM previews, Nemotron integrations for SQL Server 2025 and tooling to onboard enterprise artificial intelligence agents into Microsoft 365.
Google unveils Gemini 3 with generative interfaces and agent
Google introduced Gemini 3, a multimodal Artificial Intelligence upgrade that creates visual, interactive outputs and an experimental agent to manage multi-step tasks. The model ties deeper into search, shopping, and a new single-prompt development platform.
3D DRAM accelerator promises major artificial intelligence inference gains
d-Matrix and Alchip have announced a collaboration to build what they call the world’s first 3D DRAM-based datacenter inference accelerator, aiming to overcome memory bandwidth and cost limits in current Artificial intelligence infrastructure.
Gemini 3: Google DeepMind’s most intelligent Artificial Intelligence model
Gemini 3 is Google DeepMind’s most intelligent Artificial Intelligence model, combining advanced reasoning, native multimodality, and long-context understanding to help people learn, build, and plan. It is available via Gemini, Google AI Studio, the Gemini API, and integrations with developer platforms.
Microsoft making custom chips for Artificial Intelligence
Microsoft has developed Azure Maia 100 GPU and Cobalt 100 CPU as its first custom silicon, aimed at optimizing Azure data centers for Artificial Intelligence and cloud workloads.
Gemini 3 Pro becomes Amp’s default smart agent
The Artificial Intelligence model Gemini 3 Pro is now the primary model powering Amp’s smart agent mode, replacing Claude as the main in-product model. The update brings measurable performance gains and a raft of product features and mode changes across Amp.
d-Matrix and Alchip unveil 3D-stacked DRAM accelerator for Artificial Intelligence inference
d-Matrix and Alchip are collaborating to build the first 3D-stacked DRAM datacenter inference accelerator, promising up to 10× faster inference for Artificial Intelligence workloads and lower cost compared with HBM4-based systems.
NVIDIA Apollo: open Artificial Intelligence physics models for scientific simulation
NVIDIA introduced Apollo, a family of open Artificial Intelligence physics models to accelerate industrial and computational engineering at SC25. The models provide pretrained checkpoints, reference workflows and real-time simulation capabilities across domains from semiconductors to weather and fluid dynamics.
NVIDIA accelerates materials discovery with Artificial Intelligence and GPU computing
At SC25, NVIDIA introduced Artificial Intelligence microservices and Holoscan-accelerated pipelines to speed nanoscale imaging and large-scale molecular screening for materials research.