Perplexity serves Qwen3 235B models on Nvidia GB200 racks, showing major inference gains - Crypto Briefing
8/10Perplexity is serving Qwen3 235B large language models using Nvidia GB200 GPU racks, achieving major inference performance gains. This development was reported on May 14, 2026, highlighting Perplexity's infrastructure advancements to boost model serving efficiency.
