Yhn HealthFitness first seen 7 h ago, last 19 min ago, peak #9

Qwen3.8 hits 255 tokens per second on one RTX 5090

Original: Running Qwen3.8 at 255 tok/s on a single 5090

A developer reports running the Qwen3.8 language model at 255 tokens per second on a single Nvidia RTX 5090 GPU. The result is drawing attention from AI enthusiasts interested in high-throughput local inference, as it shows consumer-grade hardware can serve small open-source models at speeds suitable for responsive, real-time applications without data-centre equipment.

Why now: Local AI inference performance on consumer hardware is a popular topic, and this speed figure is notably high for a single GPU.

Qwen3.8Nvidia RTX 5090MeshApplied

Open on hn →

Rank over time, top of the chart is #1. 2 snapshots from 1 h ago to 19 min ago.

Evidence

• Running Qwen3.8 at 255 tok/s on a single 5090 · Ir0n · 12

API: https://socialmediatrends-api.osmike.com/v1/trends/2164021