Dev claims ~700 tokens/s running Qwen 3 27B on a single RTX 5090
IgorCarron · x · 2026-11-11
Developer @ashxhart reports hitting roughly 700 tokens per second running Qwen 3 27B on a single RTX 5090, sharing benchmark output as proof. Igor Carron amplified the result, noting how fast local inference performance keeps climbing. If verified, this shows consumer single-GPU setups with aggressive quantization and an optimized inference stack can approach server-grade throughput; implementation details are not yet published.
More from Infra
- Stratechery: Intel vs AMD's diverging earnings and Intel's need for an old-new path — Stratechery · 2026-11-11
- Alibaba files to list chip-design unit T-Head in Hong Kong to fund AI push — pstAsiatech · 2026-11-11
- AWS CEO expects to be short on compute 'for the next couple of years' as demand outstrips supply — firstadopter · 2026-11-11
- Natural language search over autonomous driving data with Cosmos-Embed1 + OpenSearch — 4310sy · 2026-11-11
- Fully local MiniMax H3 Image + LTX-2.5 pipeline generates a shot in under 5 minutes on one RTX 5090 — Hailuo_AI · 2026-11-11
- TSMC Co-COO Kevin Zhang: demand for leading-node chips is 'insatiable' — firstadopter · 2026-11-11