As of: 2026-11-22 13:32 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/p/x-metr_evals-2123864981506798012/ # “We’ve been working on understanding model capabilities beyond what our prior time…” METR (@METR_Evals), X, 2026-11-20. Source: https://x.com/METR_Evals/status/2123864981506798012 ## Archived text > We've been working on understanding model capabilities beyond what our prior time horizon tasks could capture. > > Today we’re sharing a research update. We find that AI R&D-relevant capabilities are improving rapidly and are getting harder to measure. Archived 2026-11-22 via fxtwitter (unofficial). Counts: 208,635 views, 1,582 likes, 219 reposts, 43 replies (at fetch time) Media: https://pbs.twimg.com/media/HXl78dBa4AA8nq9.jpg?name=orig ## Cited in - 2026-11-20: [METR Q4: ~40-hour horizons, judgment lags](https://postcutoff.com/e/2026-11-20-metr-q4-capability-update-time-horizon-judgment/)