Local AI server rebuild with custom loop: 70GB VRAM, load temps drop to mid-40s
Cleric07 · reddit · 2026-09-10
A Reddit user rebuilt their local AI inference server with a custom water-cooling loop: 5950X, 64GB DDR4, 2x RTX Titan 24GB and a 22GB-modded 2080 Ti for 70GB total VRAM. Load temps dropped from the upper 80s to mid-40s °C, idling at 30°C — a useful cooling reference for multi-GPU local deployments.
More from Infra
- You can now write CUDA kernels in plain Rust via cuda-oxide and cutile-rs — rjurney · 2026-09-10
- 2,400 experiments show layer dropout can match dense baselines in LLM pretraining — burkov · 2026-09-10
- Deep-Dive Speculative Decoding Blog Incoming: Drafter Training to vLLM Serving — auto_grad_ · 2026-09-10
- KV cache exposes agent economics: devs pay big for context re-reads that cost providers nothing — hackgoofer · 2026-09-10
- AMD details agent-native ROCm 10 and Hyperloom, auto-optimizing 14,000 models — AnushElangovan · 2026-09-10
- NASA chief backs orbital AI compute as SpaceX targets first space data center in 2027 — rohanpaul_ai · 2026-09-10