Ex-OpenAI dev tries GPT-5.6 Luna on Extra High: can't even find a link in the conversation
yuntiandeng · x · 2026-09-20
After running out of Codex usage, yuntiandeng tried GPT-5.6 Luna on the Extra High setting and was frustrated: the model couldn't even locate a link within their own conversation. He contrasts it with the awe he felt when GPT-3 could write code at all, wondering whether he's simply spoiled or whether Luna is genuinely that weak.
More from Models
- User burns 100M tokens on Jev without exhausting the $5 free credit — DeryaTR_ · 2026-09-20
- Users slam Codex 'reset culture': paid subscribers hoard usage waiting for free resets — TheMoonMidas · 2026-09-20
- Andrew Chen: 400x cheaper inference could unlock ad-supported free AI-native apps — andrewchen · 2026-09-20
- Bilibili's AI Infinite Arena: 190 videos stress-testing 100+ models on real tasks — vista8 · 2026-09-20
- Users report suspiciously generous usage: full morning of building leaves 78% quota left — TheMoonMidas · 2026-09-20
- Researcher flags severe LLM degradation as a mission-critical concern — doodlestein · 2026-09-20