Can 2x3090 plus 128GB RAM run 100B-class LLMs at low quants?
Borkato · reddit · 2026-10-23
A Reddit user asks whether upgrading from 48GB DDR4 to 128GB DDR4 alongside 2x RTX 3090 (48GB VRAM) would let them run DeepSeek V4.1 and GLM 5.3 Flash at low quantizations at 20 tokens/s, outperforming Qwen 3.8 Flash Next — or whether a DDR5 platform rebuild makes more sense. No answers yet.
More from Infra
- Intel CTO: chip output can at best 2X while AI compute grows 3X per year — SammaVaco · 2026-10-23
- Engineer reverse-engineers projector TV's 10s wake lag to a cold-booting Linux board — banteg · 2026-10-23
- Ollama CEO says enterprise AI is having its PC moment on This Week in AI — ollama · 2026-10-23
- Modal open-sources Spindle, a Tinker-compatible training server that charges GPU time instead of tokens — josh_wills · 2026-10-23
- Aleph Alpha documents RL sandbox infrastructure lessons: stalled GPUs, full disks, crash-as-wrong-answer bugs — _ScottCondron · 2026-10-23
- Treat proprietary silicon as strategic lock-in, architect warns — DavidLinthicum · 2026-10-23