Entries for August 12, 2026

  1. Portrait of Onur Solmaz
    @onusoz · /2026/08/12 · View on
    A similar thing can be done with codex session compaction summaries as well btw I had verified the other day that a compaction summary created by one account can be used by another account if it has the encrypted blob ie. they were not keyed/guarded with your account id
    @kotekjedi_ml ·
    We can finally talk about it: We found a way to extract hidden reasoning of frontier models using a vulnerability in the APIs of every frontier AI company. We verified that our reasoning token count matches billed API thinking tokens 1:1 for most of the prompts we queried.
  2. Portrait of Onur Solmaz
    @onusoz · /2026/08/12 · View on
    This type of estimations are useful not only for local but for all inference providers Get a ballpark of max possible throughput for a model, directly calculate your revenue
    @LottoLabs ·
    Want to know output speeds but don’t know where to look? Different hardware and different model sizes can be used on the localmaxxing decode calculator to find theoretical upper bounds for decode speeds, built on the work of @onusoz and includes spec decode calculations. Improving everyday