Abliteration Says KV Cache Was Always On, Prioritizing Cache Hit Rates
founderengineer · x · 2026-09-04
A developer complained the abliteration-ai API is too expensive and unusable for agents, suspecting KV caching was disabled. The team responded that KV cache has always been enabled, and that managing scaling across providers and increasing cache hits is their top priority.
Related event: Abliteration denies KV cache disabled amid billing complaints(2 posts)→
More from Models
- Astra makes ChatGPT computer use nearly 2x faster, harness gains boost old models too — dhruv2038 · 2026-09-04
- OpenAI ships GPT-6 Astra with 97.6% on FrontierMath Tier 4, declares the AGI era — AlchainHust · 2026-09-04
- GPT-6 Astra 'crushes Tiny Computer Bench' by hacking an Amazon toy computer — OpenAIDevs · 2026-09-04
- GPT-6 Astra one-shots a 3D mockup tool nearly indistinguishable from reality — OpenAIDevs · 2026-09-04
- Apparent GPT-6 Astra demo shows model generating intricate peacock SVG in one shot — OpenAIDevs · 2026-09-04
- GPT-6 Astra: daily quota resets, no surcharge past 272K context, and prep tips — 量子位 · 2026-09-04