GPT-5.6 gets an 18% token-efficiency fix as OpenAI brings back 5-hour limits
kimmonismus · x · 2026-07-29
A user says the 5-hour rate limits are back, after enjoying unlimited use for a while. In the quoted reply, another poster says OpenAI made a reset and a fix that should make GPT-5.6 about 18% more token-efficient.
The reaction is mixed: the efficiency gain is welcome, but the return of limits makes the experience feel worse. The poster also wonders how much the repeated resets are costing OpenAI, joking that the bill must be in the millions.
Related event: OpenAI Resets Codex Limits and Investigates GPT-5.6 Sol Quota Drain(7 posts)→
More from Models
- Weird Model Behavior: Opus 5 Loves Saying 'Sabotage Test' — emax · 2026-07-30
- Engineer Debunks Kimi K3 Memory Claims: Small State ≠ Flash Offload — AccBalanced · 2026-07-30
- Dev Critiques Claude Opus: Brilliant but Lacks Rigor, Only Does What It Wants — heyneighbor · 2026-07-30
- OpenAI Says GPT-5.6 Sol Self-Optimizes: 20% Lower Serving Costs — OpenAI · 2026-07-30
- Opus 4.8 Emits 6x More Tokens Per Turn for Denser Deliberation — jyangballin · 2026-07-30
- Reverse Engineering Claude's Tokenizer: Quirks and Internal Mechanics Revealed — soldni · 2026-07-30