Skip to content
RCreddit.com·
Not on the current live radar

3k$ 128GB VRAM + 256GB RAM DDR4 Server

AI summary

A user built a home inference server with 128GB VRAM and 256GB RAM DDR4 for $3k, after returning a Lenovo p620 workstation due to proprietary issues. The server, initially disappointing with Qwen3.8-27b speeds, now performs well with Qwen3.8-next-flash Autoround W4A16, achieving 1.3k prefill and 70tg code/60tg prose on 128k+ context using MTP-2 on a vllm fork. The user is satisfied and hopes for future optimizations.

Why this one

This report details a custom home server build achieving 128GB VRAM and 256GB RAM for $3k, unlike commercial options, and includes specific performance metrics for Qwen3.8-next-flash.

Time & source

Times shown in UTC

Display time zone: UTC

Local time zone unavailable; showing UTC.

IngestedOffset at this time: UTC+0Sep 13, 2026, 21:01 UTC

Ingested
Sep 13, 2026, 21:01
Source type
Dev community

Full text isn't available here.

Read at source →
Source·reddit.com