page_tag_header_title

OpenAI says Jalapeño beat Nvidia GB300 in two inference benchmarks at 700 watts, with deployment planned later in 2026 to cut costs and speed responses. Key Points: OpenAI says Jalapeño led Nvidia’s GB300 in throughput per unit of power and response speed during testing. The 700-watt inference chip

Elon Musk said SpaceX and Nvidia plan a 2027 Vera Rubin orbital launch, but both stocks fell, with Nvidia down 2.91% and SpaceX closing at $135. Key Points: SpaceX and Nvidia are targeting Q4 2027 for the first space-optimized Vera Rubin NVL72 launch, with broader scaling planned in 2028. Nvidia clo

Key Points Nvidia said its Groq 3 LPX inference accelerator entered full production. The product targets workloads from AI agents that complete multistep tasks. Nvidia is expanding its inference hardware lineup beyond model training systems. Data center power constraints remain a central issue for A

Nvidia cut its planned OpenAI Ohio guarantee below $120 billion from a previously discussed $250 billion. Key Points: Nvidia is expected to guarantee less than $120 billion initially, down from a previously discussed $250 billion. The chipmaker would back only the first phase, while OpenAI is still

Nvidia is working with six financial groups to mobilize more than $500 billion for AI infrastructure, deepening Wall Street’s role in funding the global compute expansion. Key Points: Nvidia plans financing platforms designed to mobilize more than $500 billion in third-party capital for AI infrastru