$NVDA···Nvidia cuts Rubin Ultra HBM from 1TB to 192GB as supply constraints force bandwidth-over-capacity designmyron·SemiAnalysis·11 days agoshow more ▾
$VRT···Vertiv expanding addressable content per megawatt via modular power packagesnico·SemiAnalysis·14 days agoshow more ▾
$SPCX···SpaceX targets 10GW data center capacity by 2027 to capture $300B ARR from AI token economicsjeremy·SemiAnalysis·2 months agoshow more ▾
$GOOGL···Google talent exodus signals innovation decline but stock benefits near-termdoug·SemiAnalysis·2 months agoshow more ▾
$SKHY···SK Hynix misses earnings on LTA shift as memory rate of change deceleratesdoug o'loughlin·SemiAnalysis·2 months agoshow more ▾
$CRWV···CoreWeave pioneered GPU project finance with Blackstone, now benchmark for neocloud debt pricingdan·SemiAnalysis·2 months agoshow more ▾
$SPCX···xAI's compute rental with clawback clauses is a valid RSI financing strategymax·SemiAnalysis·2 months agoshow more ▾
$MOONSHOT-AIMoonshot's Kimi K3 ranks third globally behind GPT-4.5 and Sonnet 3.6max·SemiAnalysis·2 months agoshow more ▾
$RCAIRCAI CTO explains pivot to pre-training for developer sovereigntylucas atkins·SemiAnalysis·2 months agoshow more ▾
$PEBBLEPebble optimizes GPU power-performance curves to boost tokens per watt and eyes 100 GW flexible power marketkav ul shah·SemiAnalysis·3 months agoshow more ▾
$RADIANTRadiant launches as neo-cloud with vertical integration, targeting 12-month greenfield-to-production timelinepatrick o'shaughnessy·SemiAnalysis·4 months agoshow more ▾
$AMD···AMD Infinity Fabric on-package interconnects precedent for CPO integrationunknown·SemiAnalysis·4 months agoshow more ▾
$ANTHROPICAnthropic secures massive compute via Colossus 1 rental to lift Claude token limitsunknown·SemiAnalysis·4 months agoshow more ▾
$AMZN···AWS Trainium 3 delivers 30-40% better price-performance, scaling to 1M+ chips in 2026randa·SemiAnalysis·5 months agoshow more ▾
$POSITRONPositron bets on LPDDR memory and 1:1 matrix-vector ratio to slash inference coststhomas summers·SemiAnalysis·5 months agoshow more ▾
$NVDA···SemiAnalysis benchmarks confirm Nvidia inference leadership across cost and throughputbryan shan·SemiAnalysis·6 months agoshow more ▾
$QCOM···Qualcomm AI 200 targets inference with LPDDR5X but memory costs rising; AI 250 compute-near-memory could be worthwhiledylan patel·SemiAnalysis·7 months agoshow more ▾
$SPCX···XAI rapidly scales gigawatt AI infrastructure across two Memphis sitesunknown·SemiAnalysis·9 months agoshow more ▾