Upgrading a Local LLM Server: Build Discussion
CryMoreT_T · reddit · 2026-07-17
The author discusses upgrading their local LLM server, aiming for better cost-effectiveness within a budget of $3,000–$3,500 while ensuring future scalability.
Current specs include:
- i9-12900K
- Z690 motherboard
- 32GB DDR5
- 3090 + 3060 + 2080 Ti with 22GB VRAM
The plan is to transition to a server/workstation-grade setup:
- ASRock Rack ROMED8-2T/BCM motherboard
- EPYC 7302P
- 512GB DDR4 ECC
- 12 GPU mining rig frame + 3x 1600W power supplies
- Target total VRAM: 244GB
They also mention wanting to run the maximum context for DeepSeek v4 flash, considering q8, or running lower-quant versions of GLM5.2 or Kimi K3.
More from Infra
- Tesla’s FSD v14 Lite is reportedly headed to 4 million older HW3 cars — MatthewBerman · 2026-07-21
- TSMC’s 3nm utilization reportedly tops 120% as AI demand drives a $190B capex cycle — tengyanAI · 2026-07-21
- Nativ brings local AI model running to Mac with a desktop app and localhost API — Simon Willison · 2026-07-21
- Octen says agent search now runs at 62ms P50 with only a 6ms P90 gap — aakashgupta · 2026-07-21
- Zhipu acquires a compiler-team spinout to optimize AI inference on domestic chips — zephyr_z9 · 2026-07-21
- Open reproduction of Meta’s REWIRE data pipeline cuts the cost to about $11 — vanstriendaniel · 2026-07-21