AI R&D-relevant model capabilities are improving rapidly and are getting harder to measure METR's capability evaluations of public models up to Opus 5.5 and GPT-6 Astra find that AI R&D-relevant capabilities are improving rapidly and are getting harder to measure, and that current public evidence is insufficient to rule out drastic AI R&D acceleration within the next couple of years. METR defines drastic acceleration as capabilities improving so fast that humans can no longer meaningfully understand, assess and react to catastrophic risks from new AI systems. The post also reports that agent swarms appear helpful as an elicitation technique on verifiable tasks with a clear metric and fast feedback compared with best-of-k, but have unclear or negative effects on METR's other tasks. This post presents evidence from our capability evaluations of public models relevant to AI R&D acceleration, up to Opus 5.5 and GPT-6 Astra. We remain significantly uncertain about central questions, including 1 what capabilities are in practice necessary for drastic acceleration, and 2 how quickly relevant model capabilities are improving. We think that current public evidence is insufficient to rule out drastic AI R&D acceleration in the next couple of years. By drastic AI R&D acceleration we mean a situation where capabilities improve so quickly that humans are no longer able to meaningfully understand, assess, and react to catastrophic risks posed by new AI systems. This overview summarizes our main takeaways and their relevant sections within the post. 1 Background 1-background describes our motivation and aims, and how our thinking about capability evaluations has changed since the release of the original time horizons paper https://metr.org/time-horizons/ . 2 Methodology 2-methodology describes the different task sets used throughout the post, changes to human baselining methodology, and introduces our use of multiple score thresholds per task. 3 Results 3-results contains the evidence for our main takeaways, which are: 4 Robustness 4-robustness-checks details various checks we did on our results, such as checking for reward hacking and refusals. We also investigate the effectiveness of agent swarms as an elicitation technique and find that they seem helpful on verifiable tasks with a clear metric and fast feedback compared to best-of-k