Rig Finder index / 8B / Casual chat / California
Best rig to run an 8B model for casual chat in California
At this usage, renting Claude Sonnet 4.6 beats owning for every rig that runs an 8B model. The least-bad build is the Mini-PC, but that's a sovereignty call, not a savings one.
Same workload, same cloud, same electricity — we swap only the hardware and rank by real break-even. Rigs that can't run an 8B model are shown, never recommended. Compared against Claude Sonnet 4.6 ($3/M in, $15/M out); change the cloud in the lab.
Every build, ranked
| Rig | Build | 8B tok/s | Local/mo | Break-even | Verdict |
|---|---|---|---|---|---|
| Mini-PCbest payoff | $750 | 22estimated | $22.10 | ~38 mo | RENT |
| RTX 3090 (used) | $1,200 | 85estimated | $39.13 | ~79 mo | RENT |
| Mac M-series | $3,000 | 65estimated | $84.49 | ~151 mo | RENT |
| RTX 4090 | $2,300 | 120estimated | $70.31 | ~158 mo | RENT |
| 2× RTX 3090 | $2,250 | 90estimated | $72.35 | ~202 mo | RENT |
| RTX 5090 | $3,500 | 180estimated | $105 | ~270 mo | RENT |
Assumes casual chat use: ~3M tokens/month, 1h/day active, 3h/day idle, at California electricity ($0.322/kWh). Speeds are our estimates; community-measured tok/s live on the interactive Rig Finder and the benchmarks board. Every formula is on the methodology page.
Run this with your exact numbers →