GPT-5.6 Pro appears to downgrade to GPT-5.5 Mini across web, desktop and iOS
yuchaocheung-pixel · ghdev · 2026-07-22
GPT-5.6 Pro appears to reroute to GPT-5.5 Mini across web, desktop, and iOS
A GitHub issue against openai/codex reports a possible silent routing mismatch: after selecting GPT-5.6 Pro, the account consistently gets Instant-like responses with no visible reasoning, and the model reportedly identifies itself as GPT-5.5 Mini.
- Reproduced on ChatGPT Web for Windows, Codex Desktop, and ChatGPT for iOS
- The reporter says the behavior is consistent even in a private/incognito browser session
- The issue asks OpenAI to inspect requested-model vs served-model routing, account eligibility, and any safety flags
- If the downgrade is intentional, the user wants the UI to disclose the actual served model
More from Models
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11
- Terminal Bench v4: GLM-5.3 Leads at 41.9%, Kimi-K3 Underwhelms at 12.6% — Ok_Warning2146 · 2026-09-11