Paychecks Calculator
Calculate Your Take-Home Pay After Taxes

Estimate the real cost of self-hosting an AI model, line by line.

How Much To Run AI is a set of free cost calculators and model guides for people budgeting to run AI models. For supported open-weight text models such as Kimi, GLM, DeepSeek, Qwen, Llama and Gemma, it estimates the VRAM needed, matches hardware, and compares local hardware with cloud GPU rental. Separate calculators cover image, video and decision models. Results are planning estimates, not vendor quotes or performance guarantees.
Each price reference shows its source type, update time, region and billing unit, and estimates without a verifiable link are marked as such.
The site publishes guides for individual hosted API, open-weight and restricted-preview models, covering access routes, license notes and cost considerations such as token rates.
Teams deciding between self-hosting and a managed API can compare upfront spend, monthly operating cost and break-even timing.
Engineers can see how model size and precision change the VRAM requirement and GPU tier.
The cost calculators are free to use. The pricing page lists generation credit packs but states they are currently unavailable for purchase.
Application development, fine-tuning, security and compliance are not included, nor are reserved capacity, negotiated rates, volume discounts or resale value.
Submit your product for free with a PickApps badge, or explore priority review and sponsored placement options. All submissions are reviewed.
Submit your product