Compare the total cost of owning a GPU and self-hosting an open-weights model against paying a cloud API. Hardware, electricity, cooling, and break-even included.
100% free · no signup · browser-local
Share on XWhat you pay a provider today.
Hardware, power, and ops.
Cloud monthly
$2.70
API cost at this volume
Local monthly
$69.74
hardware + power + ops
Monthly savings
$67.04
cloud is cheaper
Break-even volume
258.3M
tokens/month for local to win
Throughput estimate is illustrative for a 7B-class model on this GPU. Larger models reduce throughput significantly.
Cloud is cheaper at this volume. Local only pays off at higher scale.
Important caveats
Estimates are illustrative. Throughput figures assume a 7B-class model; larger models run slower. Confirm hardware prices and electricity rates for your region.
Model multi-step agents with tool calls, retries, and monthly run volume.
Compare 100% frontier traffic against a light → general → complex model cascade.
Calculate GGUF weights + KV cache, match RTX / Apple Silicon / Strix Halo rigs, and compare Ollama, llama.cpp, MLX & vLLM.
We'll email you when model prices change or new tools launch. Unsubscribe anytime.