OpenAI está probando “Luna Reserve”, un nivel de respaldo de Codex
La función no documentada da a algunos usuarios Plus y Pro una asignación separada después de que se restringe el acceso a modelos avanzados. RuntimeWire rastreó la lógica de conmutación y medición de modelos del cliente de Windows de OpenAI.
By Ryan Merket · Published · Updated
RUNTIMEWIRE INVESTIGATION — Scoop
Original reporting by RuntimeWire, based on reverse engineering.
Por qué importa
Reserve permitiría a OpenAI reemplazar el límite estricto de uso de Codex por una alternativa más barata, manteniendo a los desarrolladores activos mientras convierte la capacidad premium en una venta adicional inmediata.
Reporting record
Finding
Codex contains a feature-gated “Luna Reserve” allowance that appears to switch eligible Plus and Pro users to a Luna-class reserve model after advanced-model limits are reached.
How we verified
Methods: reverse engineering.
Static analysis of the production-distributed Codex app.asar identified the model identifier gpt-reserve, feature gate reserve_enabled, luna_reserve banner logic, dedicated lifecycle analytics, eligibility checks for Plus/Pro accounts, and user-facing text describing a separate “Luna Reserve” allowance. The code preserves the user’s original model selection and restores it when Reserve ends.
Calculated the archive’s SHA-256 hash, parsed its ASAR index, and extracted the Electron main-process and renderer bundles. Searched the bundles for feature definitions, model identifiers, rate-limit handling, eligibility conditions, analytics events, and user-facing strings. Traced Luna Reserve from its server-controlled feature gate through account eligibility, rate-limit activation, model switching, UI presentation, and restoration of the original model. Compared findings against current official OpenAI documentation. No server interaction or feature-flag bypass was attempted.
Tested versions:
- Windows package 26.818.5229.0
- embedded Codex app 26.818.41509
- build 6962 (prod)
Reproduction
RuntimeWire independently reproduced the core finding.
Obtain the specified Microsoft Store build of Codex and hash app/resources/app.asar. Parse the ASAR and inspect the renderer bundle for gpt-reserve, reserve_enabled, luna_reserve, and CodexLunaReserveLifecycle. Confirm the surrounding code includes Plus/Pro eligibility checks, a separate gpt-reserve rate limit, model switching, and the user-facing string “Luna Reserve · {remaining}% remaining.” Live UI reproduction additionally requires server-side eligibility and rate-limit state.
File hashes
C5D839BC9B122B7EF2A2F0F45186B3E5895923DE5B6CEF5253C936FE670C0479
Company response
RuntimeWire requested comment; the company had not responded by publication time.
OpenAI ha comenzado a desplegar una capa de reserva no documentada para Codex que puede mantener a los suscriptores trabajando después de que su asignación regular de modelos avanzados se agote.
La función se llama Luna Reserve. Lleva el identificador de modelo interno gpt-reserve, mantiene su propia asignación de uso y está diseñada para reemplazar temporalmente el modelo seleccionado por un usuario mientras preserva las configuraciones necesarias para revertir el cambio más tarde.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.