Sunk Cost
How long until a local LLM rig pays for itself?
Sunk Cost estimates how long a machine for running local AI models takes to pay for itself compared with paying for an API. Pick a Mac mini, Mac Studio, DGX Spark or Strix Halo box, see which open models fit and how fast they run, and adjust your usage, electricity price and API assumptions. First shared by its maker on Show HN: https://news.ycombinator.com/item?id=49706656
✨ The overview, FAQ and details below were drafted by AI from the product's own website and haven't been reviewed by the maker yet. Are you the maker? Claim this listing to review and edit them.
Overview
Who is it for?
Developers and teams deciding whether to buy hardware for local LLMs.
Problem
It's hard to know whether a local AI machine will be faster, capable enough or cheaper than an API for your usage.
Solution
A break-even calculator with model fit, estimated speed and cost based on your hardware and daily token use.
What makes it unique
It labels estimated speeds as estimates and lets you assume API prices keep falling.
Sunk Cost FAQ
Which machines can I compare?
Mac mini, Mac Studio, DGX Spark and Strix Halo systems, with editable specs.
How is local speed estimated?
Where nothing has been measured, from memory bandwidth divided by bytes read per token, and labeled as an estimate.
Does it track me?
Your inputs are sent once you stop typing, without cookies or IP address, only the request's country. Browsers set to Do Not Track send nothing.
Details
- Pricing
- Free
- Categories
- LLMs & InfrastructureResearch & Data
- Built for
- DevelopersFounders
- Platforms
- Web
Hunted by AI Launch Team
Discussion (0)
Sign in to join the discussion.