Leak claims OpenAI's Astra post-training unfinished, can't yet convert compute into gains
teortaxesTex · x · 2026-09-04
An unverified claim on X suggests GPT-6 Astra's post-training is far from done: the model reportedly cannot yet reliably and monotonically turn more compute into better results, hinting at an unstable scaling curve. The post notes Anthropic's Fable camp disagrees.
More from Models
- OpenAI launches GPT-6 Astra, claiming it can do anything you do on a computer — kagigz · 2026-09-04
- Databricks evals: GPT-6 Astra claims SOTA on OfficeQA Pro benchmarks, cheaper per task — downingARK · 2026-09-04
- Mathematician tests GPT-6 Astra: live Lean proof verification while writing arguments — teortaxesTex · 2026-09-04
- Tavus Launches Sparrow-2, Claiming #1 in End-of-Turn Detection and Interruption Handling — ycombinator · 2026-09-04
- ARC-AGI-3: 10x reasoning tokens cuts total cost from $48k to $26k vs medium — i_dg23 · 2026-09-04
- Researchers flag data contamination concerns in benchmark behind Astra's time-horizon score — dfrsrchtwts · 2026-09-04