← AI Hashrate δΈ­ζ–‡

Local vs Cloud Cost

Amortize hardware over 3 years, add electricity, and compare per-1M-token cost against major cloud APIs. Pick the view that matches your usage. Methodology.

GPU Setup cost.col.vram cost.col.tps Annual Cost Tokens/yr $/1M tok GPT-4oGPT-4o-miniClaude Sonnet 4Claude Haiku 4DeepSeek V3Qwen3-8B (Together) Break-even
Arc B580 12GB12GB35.4 tok/s$166/yr372.1M/yr$0.45$6.25$0.38$9.00$2.40$0.69$0.101628d
RTX 4060 Ti 16GB16GB22.4 tok/s$238/yr235.5M/yr$1.01$6.25$0.38$9.00$2.40$0.69$0.103689d
RTX 4090 24GB24GB78.2 tok/s$730/yr822.0M/yr$0.89$6.25$0.38$9.00$2.40$0.69$0.103241d
RTX 3090 24GB24GB72.6 tok/s$653/yr763.2M/yr$0.86$6.25$0.38$9.00$2.40$0.69$0.103123d
Mac mini M4 32GB32GB9.3 tok/s$401/yr97.8M/yr$4.10$6.25$0.38$9.00$2.40$0.69$0.1014972d

Assumptions: 8h/day usage, $0.15/kWh electricity, 3-year amortization. Cloud prices approximate mid-2026. Tok/s estimates at Q4/8K. Actual costs vary.