OpenAI Asks Users to Report Astra Bad Threads via /feedback Ahead of 6.1
ivan_bezdomny · x · 2026-09-12
OpenAI staffer pvncher asked users who notice Astra misbehaving to submit pertinent threads via /feedback — the team monitors submissions directly, no need to reply with IDs. Developer theo confirmed the OpenAI team takes feedback seriously and will fix reported threads before the 6.1 release.
More from Models
- OpenAI adds Astra-generated showcases and games to its developer site — Dimillian · 2026-09-12
- DeepSeek's edge: each model iteration targets the last one's biggest bottleneck — xennygrimmato_ · 2026-09-12
- Kimi K2.7 RL recap: gains transfer to unseen benchmarks while steps drop ~35% — echen · 2026-09-12
- RL on just 1,700 tasks lifts Kimi K2.7 across five coding benchmarks — echen · 2026-09-12
- User vibes: fable 5.1 'feels alive' while astra feels like a soulless ultrabot — rudrank · 2026-09-12
- Benchmarks broken: pass/fail scores hide models outsmarting strict hidden tests — himanshustwts · 2026-09-12