Live page · Day archive

Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s

Read the original at Hacker News

Summary

Niko1221 releases Strata, an inference engine that runs the 125B Qwen 3.8 Flash Next model on consumer hardware at 100 tokens per second.

Carried by: Hacker News. First seen: .