Poolside Laguna S 2.1 hits 256K context on a 96GB local GPU rig
_TheWolfOfWalmart_ · reddit · 2026-07-22
The author says Poolside’s Laguna S 2.1 arrived just in time for a newly upgraded local rig: 3× V620 GPUs for 96 GB of VRAM.
- They’re testing Laguna with Q4KM quantization and 256K context at F16.
- Early results on a Dell PowerEdge R740 with dual Xeon Gold 6248R and 768 GB RAM: about 400–600 tok/s prefill and 16–20 tok/s generation so far.
- They note dflash is still disabled, and the setup is costing about $350 per card.
- The user is running an HTML flight-simulator test to probe real-world behavior.
More from Infra
- Cloud capacity crisis is now an AI architecture problem, argues new newsletter — DavidLinthicum · 2026-10-26
- SpaceX files $40B GPU financing commitment with Apollo, already at 1.7GW — XFreeze · 2026-10-26
- Meta plans to raise up to $95B to fund massive AI expansion — Polymarket · 2026-10-26
- Nebius CEO: Debt Markets Closed for Neoclouds Last Week, Leaning on Customer Advances — zephyr_z9 · 2026-10-26
- CoreWeave 5-year CDS spikes to ~1,100bps as market fears AI bubble starts as credit event — kimmonismus · 2026-10-26
- AI demand reshapes cloud computing: scarce capacity and long-term deals replace on-demand — DavidLinthicum · 2026-10-26