Dev ditches Codex: 'GLM 5.3 Flash and DeepSeek V4.1 Flash is all you need'

TheZachMueller · x · 2026-09-17

Developer TheAhmadOsman says he has stopped using OpenAI's Codex models, calling them slow and extremely limited, and now relies on GLM 5.3 Flash and DeepSeek V4.1 Flash for coding work instead. The post resonated widely, signaling that cheap, fast open-weight models are displacing closed coding models for many developers.

Original post →

More from Models

Models channel →