Users Complain Codex Is Too Slow to Even Use Quotas

Multiple Reddit users report OpenAI Codex has become painfully slow, with Astra at 13.8 tok/s and Sol 6.1 at just 9.7 tok/s in new sessions, leaving quotas unused because users can't wait for responses.

2026-10-13 ~ 2026-10-14 · 2 related posts

• Codex speeds hit new lows: Astra at 13.8 tok/s, Sol 6.1 at 9.7 tok/s — rgb328 · 2026-10-13
• Codex users complain it's so slow they can't even burn through their quota — k0rtes · 2026-10-14