Perplexity CEO: inference now served on NVLink Blackwells, Vera Rubin next
AravSrinivas · x · 2026-09-09
Perplexity CEO Arav Srinivas says the company is now serving much of its inference on NVIDIA NVLink Blackwells, with Vera Rubin next. The cited NVIDIA case study notes Perplexity serves 50M+ daily queries across 15 models including Nemotron 3 Ultra, and Blackwell's larger NVLink domain lets models scale across nodes without latency sacrifice. NVIDIA also released an AI Tokenomics white paper covering token demand forecasting, cost optimization, and pricing, with Cohere, Perplexity, and Canva case studies.
More from Companies & People
- Altman responds to Navier-Stokes proof feud: OpenAI offered rival team to go first — sama · 2026-09-09
- Cohere: 80% of study participants rely on non-English languages daily — Cohere_Labs · 2026-09-09
- OpenAI's Navier-Stokes claim sparks pushback from mathematicians Buckmaster and Tao — fortnow · 2026-09-09
- Researcher says allegations against Sebastian believable, cites past AI-proof incidents — BlancheMinerva · 2026-09-09
- AgentCon London wraps up with Microsoft and AWS as sponsors — lee_stott · 2026-09-09
- Former OpenAI intern vouches for '100% believable' misconduct allegations against Seb — xennygrimmato_ · 2026-09-09