News for hackers. Stuff that matters.

A comment on AirLLM 70B inference with single 4GB GPU. Parent comment

Re:AirLLM 70B inference with single 4GB GPU by jackb4040 on Monday August 03, 2026 @09:22AM (#49155481)
Ahaha thank you, I naively assumed the unlabeled graph in the readme was tps, not spt!
[ Parent | on HN ]
Re:AirLLM 70B inference with single 4GB GPU by throwawayffffas on Monday August 03, 2026 @10:23AM (#49156196)
Running 292 tps on K3 would make your gpu a money printer.
[ Parent | on HN ]
Re:AirLLM 70B inference with single 4GB GPU by meneton on Monday August 03, 2026 @12:05PM (#49157624)
spt not tps
[ Parent | on HN ]