v2.20.0

Released August 20, 2026 · View on GitHub →

GLM-5.3 reached Z.AI's pay-per-use API and now carries the rate they publish, instead of no price at all.

Added

subscribers first; the standalone model API accepts it as of 2026-08-19, so it now appears on that roster too and is its default. The China gateway (z.ai-cn, z.ai-cn-api) is a separate listing that Z.AI's page does not cover, so it keeps the GLM-5.2 roster — offering an id a gateway rejects would be a guaranteed 4xx.

Changed

by design: no per-token rate existed while the model was Coding-Plan only, and borrowing GLM-5.2's would have been an invented number. Z.AI now publishes one, and it happens to equal GLM-5.2's — a fact about today's price list, not a rule linking the two. /cost and the dashboard stop showing pay-per-use GLM-5.3 traffic as free. Coding Plan usage is flat-fee and still reads "Included in plan".

Downloads

Install with npm install -g codeep@2.20.0 or brew install vladoivankovic/codeep/codeep.