Should this founder drop €8k on a 5090 workstation or stick with cloud GPUs?
abandonedexplorer · reddit · 2026-09-15
A Reddit user is debating an €8k local workstation (RTX 5090 + 128GB RAM) for AI inference. Key points:
- Workloads: mostly image generation, video generation and animation, not LLMs; his AI side business already makes real money
- Tax angle: as a business purchase he can deduct VAT plus roughly €2k in write-offs, meaningfully lowering real cost
- Case for owning: 6-12+ hours/day usage at scale; fears of rising cloud GPU prices, worsening availability, privacy, and providers adding monitoring/KYC/restrictions on generative AI; values self-reliance
- Doubts: current GPU/RAM prices are insane while RunPod/Vast remain affordable
He's asking whether to go all-in on hardware or a cheaper local + cloud hybrid.
More from Infra
- Expert-lookahead delivers 10%+ speedup for MoE inference on low-memory Macs — carloslfu · 2026-09-15
- Local voice assistant keeps its memory across model swaps, thanks to open-source Mnemosyne — Enough_Leopard3524 · 2026-09-15
- Hugging Face reportedly acquired by NVIDIA — Turn_Trout · 2026-09-15
- NextDC accused of using AI to draft lobbying letters, with AI errors, for 225MW datacenter expansion — nordicinst · 2026-09-15
- vLLM-Omni: serving voice, video, and diffusion models explained by Red Hat AI — vllm_project · 2026-09-15
- Running Z-Image Turbo locally on an RX 6800: full ROCm setup, 43s per image — AstroFieldsGlowing · 2026-09-15