H3 model crashes at 0.5 resolution, seeking config fix
reicaden · reddit · 2026-08-24
A user faces severe performance bottlenecks running the Minimax H3 model in ComfyUI, unable to exceed 0.5 resolution and 6-7 seconds duration despite using a 5090 GPU and 128GB RAM. Even with Sage attention and easy cache nodes enabled, performance lags behind others with similar hardware, leading the author to seek optimization advice.
More from Infra
- Fluidstack hires Electrical Engineer for Modular R&D to build frontier AI infrastructure — MxMnr · 2026-10-16
- Jensen Huang joins BG2 podcast: does AI revenue justify capex, '27 bottlenecks — elonmusk · 2026-10-16
- Mistral's on-device infra rack revealed: DGX Spark, Mac Studio and more — gregmushen · 2026-10-16
- NVIDIA and Sweden's Wallenberg launch Sferical One, a shared AI supercomputer — nvidia · 2026-10-16
- Next infra frontier: self-evolving systems that adapt and self-verify — _shreya_s · 2026-10-16
- Xiaomi's MiMo RL dashboard reveals $3.47M spent on just 30 training steps — bycloud · 2026-10-16