Show HN: Getting GLM 5.2 running on my slow computer
Points and comments are a snapshot, not live.
Article body wasn't reachable. The HN discussion summary is below.
Points and comments are a snapshot, not live.
Article body wasn't reachable. The HN discussion summary is below.
What commenters are saying
Commenters widely praised the hacker spirit behind the project. Two camps emerged: those who find 0.05-0.1 tok/s usable for overnight batch tasks, and those who see it as too slow for interactive use. Multiple commenters suggested ticket-style interfaces or batch task queues as better fits for slow local inference. Some raised SSD wear concerns, while others noted read-only access minimizes that risk. Several offered to test on higher-end hardware, and antirez was credited with similar work on a GLM 5.2 branch in ds4.