Home Posts Projects

VRAM training vs inference requirements for different models

02-06-2026

Training:

  • https://docs.google.com/document/d/1J2BX9jkE5nN5EA1zYRN0lHhdCf1YkiFERc_nwiYqCOA/edit?tab=t.0#heading=h.x4gj8avqmrv3

Inference:

  • https://apxml.com/tools/vram-calculator
  • https://vgel.me/posts/faster-inference/

Model Performance:

  • https://horace.io/brrr_intro.html

Related

27-08-2024

GitHub Models and other LLM inference services

© 2026 Luke Miloszewski

Email Address GitHub Twitter