Show HN: Qwen3.8-27B API, 140 tok/s on one GPU

 (inference.tiyuvta.ai)

1 points | by anotherCodder 3 hours ago

1 comments