ASKED FOR ASTRA. GOT LUNA.

AI News Page first flagged the mismatch on September 20 after finding two public technical traces in separate GitHub repositories. Both showed a request for GPT-6 Astra followed by a successful response identifying GPT-5.6 Luna rather than Astra. RECEIPT · RECEIPT

The report in OpenAI's Codex repository recorded seven traced Astra requests across low, medium, high and xhigh reasoning. Every one returned a response object labeled GPT-5.6 Luna, while control requests for GPT-5.6 Sol, Luna, Terra and GPT-5.5 were labeled as requested. The same Astra responses carried headers saying safety buffering was enabled and naming GPT-5.6 Luna as the faster model. RECEIPT

A separate CLIProxyAPI report showed its upstream request body selecting GPT-6 Astra and the upstream response returning GPT-5.6 Luna with the same safety-buffering headers. A later OpenAI Codex issue added adapter-mediated trace samples from four account profiles and a larger local corpus where every Luna-labeled mismatch was associated with an Astra request, while explicitly warning that the adapter evidence cannot prove which model weights actually ran. RECEIPT · RECEIPT

The narrow finding is now hard to dismiss: several public captures show Codex subscription or OAuth requests asking for Astra and receiving responses reported as Luna. What remains unresolved is whether that model field always identifies the actual serving weights, how broadly the behavior occurs, and whether the substitution is an intended fallback, a routing bug or a labeling problem. OpenAI's public model documentation treats Astra and Luna as distinct models, but does not explain this observed routing behavior. RECEIPT · RECEIPT · RECEIPT