Solah retweeted
MENSAH β‘οΈ MALI
π: miamihurricanes.com/heismen
Solah retweeted
STILL NOT OVER THIS SHOT π
WHO WAS WEMBY FEELING LIKE?? teapawt.net/BleacherReport/status/β¦
Solah retweeted
βFor a time is coming when people will no longer listen to sound and wholesome teaching. They will follow their own desires and will look for teachers who will tell them whatever their itching ears want to hear. They will reject the truth and chase after myths.β
2 Timothy 4:3-4
Solah retweeted
$NVDA is building the infrastructure for the largest technological transformation in human history and still trades at a lower multiple than a paint company.
The market wonβt be able to keep $NVDA below $200 much longer.
Jensen is clearly building for the inference era with Rubin handling attention-heavy workloads and Groq powering low-latency token generation.
The AI engine just keeps humming.
RT @Sam_Badawi: Trump says he is not deploying troops anywhere.
He reiterated that he expected the war to be worse and believes it will enβ¦
This quoted post is unavailable.
Solah retweeted
$NVDA NVIDIA PLANS MAJOR ANNOUNCEMENTS ON INFERENCE AI, AGENTIC SYSTEMS, AND NEXT-GENERATION DATA CENTER ARCHITECTURE AT GTC - REUTERS
Nvidia is expected to unveil a series of AI infrastructure updates at its annual GTC developer conference as the company seeks to reinforce its leadership position amid rising competition from hyperscalers developing custom chips.
The announcements are expected to focus on inference computing and βagentic AI,β reflecting a broader industry shift as AI workloads move from large-scale model training toward real-time task execution across applications.
Nvidia may also introduce new server architectures integrating technology from its recent $17B acquisition of Groq, which specializes in high-speed inference chips designed to deliver lower latency AI workloads.
The company is additionally expected to highlight systems incorporating CPU-based infrastructure to address emerging bottlenecks in βagent orchestration,β where large numbers of AI agents coordinate tasks across software platforms.
Despite Nvidia maintaining roughly 90% market share across AI training and inference accelerators, the announcements come as customers including major AI labs and hyperscalers increasingly invest in internally developed ASIC chips for AI workloads.
reuters.com/technology/nvidiβ¦
