llmaker: Self-host a full modern LLM stack from your terminal
raiyanyahya · reddit · 2026-08-22
Developer raiyanyahya released the open-source project llmaker, which lets you self-host a modern LLM stack and run the whole fleet of services from your terminal, lowering the barrier for local LLM inference deployment.
Repo: github.com/raiyanyahya/llmaker
More from Infra
- Anthropic hires ex-TPU leader Amir Salek to develop custom AI chips — Pablomiller · 2026-08-22
- Report: Musk struck deal with ASML to purchase massive equipment — burhop · 2026-08-22
- UBS predicts $4.1T AI infra spend by 2028, ignoring power bottleneck — Servola-Journal · 2026-08-22
- Poll: 75% of Americans now oppose data centers — pstAsiatech · 2026-08-22
- Fixing MTP head cuts Ornith1.5 35B inference time by 33% — frankentriple · 2026-08-22
- PyTorch as a reference language vs. implementation language — tokenbender · 2026-08-22