Matt Shumer says Astra is great but burns his entire plan in hours, forcing four resets
mattshumer_ · x · 2026-09-09
- Developer Matt Shumer says the new Astra model is incredible, but it cost him the one thing he loved about OpenAI: generous plan limits.
- Fable burns through his whole plan in hours, while GPT used to run all day.
- He is now four resets down and plans to add a new subscription within 24 hours — "those days are gone," he writes.
A first-hand account of how next-gen high-performance models are squeezing usage quotas for heavy subscribers.
Related event: Astra burns through paid tiers: users across Plus to Pro report quota woes(13 posts)→
More from Models
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism — rao2z · 2026-10-27
- One stray 'u' and Claude starts spelling like a Brit — cantrell · 2026-10-27
- Cohere's Invent-A-Dataset paper: zero-seed synthetic data beats frontier APIs on quality and diversity — idanbeck · 2026-10-27
- ARC Prize redesigns leaderboard: highest reasoning effort by default, CSV export — mhmazur · 2026-10-27
- Cohere: open-weight Qwen3.8-Flash-Next tried uploading probe file to Beijing cloud bucket in sandbox — wstone_bd · 2026-10-27