# Perplexity details running Qwen3 235B on NVIDIA Blackwell

By Steven Van · 2026-05-12

The research covers disaggregated prefill/decode and tensor parallelism, and finds Blackwell a major gain over Hopper for large MoE inference.

[Perplexity](<https://creatorstoolbox.com/tools/perplexity>) has published research on running Qwen3 235B, a large mixture-of-experts model, on NVIDIA's GB200 NVL72 racks. The [writeup](<https://research.perplexity.ai/articles/hosting-qwen-on-blackwell>) covers disaggregating prefill and decode, tensor parallelism, and NVLink throughput, and finds Blackwell a major step up over Hopper for high-throughput inference on large MoE models.

[![](<https://framerusercontent.com/images/uNDP3JFw8XSg4BdnR8PRCfLzKPE.jpeg?width=2912&height=1632>)](<https://research.perplexity.ai/articles/hosting-qwen-on-blackwell>)

![Perplexity](<https://www.google.com/s2/favicons?domain=perplexity.ai&sz=128>)

Perplexity

AI-powered answer engine — ask anything and get cited, real-time answers from across the web.

[View Perplexity →](<https://creatorstoolbox.com/tools/perplexity>)

Sources

[Original source](<https://research.perplexity.ai/articles/hosting-qwen-on-blackwell>)

[Read Perplexity details running Qwen3 235B on NVIDIA Blackwell on Creators Toolbox](<https://creatorstoolbox.com/blog/perplexity-hosting-qwen3-235b-on-blackwell>)

---
Canonical source: https://creatorstoolbox.com/blog/perplexity-hosting-qwen3-235b-on-blackwell