A low price per unit of generated text can still produce an expensive AI job. In its June 2026 comparison, model platform OpenRouter flags Z.ai’s GLM 5.2 and MiniMax M3 as examples: their rates look attractive, but lengthy reasoning and responses can increase the bill. The useful comparison is therefore the cost of completing the same task at an acceptable quality level. OpenRouter also distinguishes hosting providers and their data policies, making the deployment choice part of the purchasing decision. Its comparison is a dated shortlist for testing, rather than a universal model ranking.
Terms in this piece · Glossary
open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
SWE-bench — The standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
