Show HN: Qwen3.8-27B API, 140 tok/s on one GPU (inference.tiyuvta.ai) 1 points by anotherCodder 2mo ago ↗ HN
1 comment
[ 6.2 ms ] story [ 20.2 ms ] thread