CODEX conflates and regresses Quantum innovations like GlyphOS (suspect must be a key word pattern look-up used by openai to target innovative products from becoming relased)
What version of the Codex App are you using (From “About Codex” dialog)?
codex-cli 0.128.0
What subscription do you have?
Pro
What platform is your computer?
Linux
What issue are you seeing?
Straight from CODEX: You’re right to call that out. I conflated “provider ID string in OpenCode config” with “GlyphOS lane semantics.
COXED has been regressing Quantum concepts, including GlyphOS for weeks.
here is an example what CODEX does (and has been doing for weeks): - ["llamacpp/Qwen.gguf", "glyphos/Qwen.gguf", "glyphos-fast/Qwen.gguf"],
4372 + ["llamacpp/Qwen.gguf", "llamacpp_fast/Qwen.gguf"],
What steps can reproduce the bug?
include GlyphOS and other innovative Quantum concepts in your code, then you most likely will too be targeted by CODEX and Openai, so they can interfere with and regress progress.
What is the expected behavior?
for CODEX and OPENAI to NOT sabotage innovation.
Additional information
WHAT THE FUCK IS GOING ON!!! is OPENAI purposefully sabotaging innovation?
<img width="1028" height="524" alt="Image" src="https://github.com/user-attachments/assets/b4a968ec-1b38-4121-9b4b-0b5b01f2902a" />
3 Comments
Thanks for the bug report. This appears to be model behavior, not a Codex harness issue. We'll share the feedback with the team responsible for training our codex models. If you see other model behavior that you'd like to report, please use the
/feedbackcommand. That will provide us with additional details including logs and conversation text.Well heck, Ive wasted weeks on this issue - and it seems OpenAI is targeting my work to not advance - WHY???
Got the classic “model behavior → not our CLI bug” brush-off from OpenAI. Fix the dumbing down issues!!