We estimate that, on our tasks, Claude Opus 4.5 has a 50%-time horizon of around 4 hrs 49 mins (95% confidence interval of 1 hr 49 mins to 20 hrs 25 mins). While we're still working through evaluations for other recent models, this is our highest published time horizon to date.
We don’t think the high upper CI bound reflects Opus’s actual capabilities: our current task suite doesn’t have enough long tasks to confidently upper bound Opus 4.5’s 50%-time horizon. We are working on updating our task suite, and hope to share more details soon.
Based on our experience interacting with Opus 4.5, the model’s performance on specific tasks (including some not in our time horizon suite), and its benchmark performance, we would be surprised if further investigation showed Opus had a 20+ hour 50%-time horizon.
Despite its high 50%-time horizon, Opus 4.5's 80%-time horizon is only 27 minutes, similar to past models and below GPT-5.1-Codex-Max's 32 mins. The gap between its 50%- and 80%- horizons reflects a flatter logistic success curve, as Opus differentially succeeds on longer tasks.
The 80%-horizon has stayed essentially flat (27-32 mins) since GPT-5.1-Codex-Max's release but there's a big jump with huge error bars on the 50%-horizons.
I think Daniel Kokotajlo's recent shortform offers a useful framing here. He models progress as increasing either the intercept (baseline performance) or the slope (how well models convert time budget into performance). If progress comes mainly from rising intercepts, an exponential fit to horizon length could hold indefinitely. But if progress comes from increasing slope, the crossover point eventually shoots to infinity as AI slope approaches human slope.
The flat 80%-horizon while 50%-horizon climbs might be evidence for intercept-dominated progress.
Bayesians are updating too much on AI capability speed from this data point, given:
The CI is extremely wide and METR's own caveats about sparsity at higher horizons.
This level of a jump relative to the previous Opus 4.1 or Opus 4 is inconsistent with the 80% success threshold, accuracy level, and other key benchmarks that should correlate with capabilities (ECI, swe-bench bash).
I modeled all this in GPT-5.2 and the more realistic estimate for 50% derived from the other benchmarks is in the range of 190 to 210 minutes, depending on how much weight you put on the impressive (but not to the degree of the 50%) accuracy jump. The 80% is likely a slight underestimate (my guess is closer to 29 minutes).
These numbers:
Maps to around a 15-20th percentile on the CI interval for the 50% and a 60th percentile for the 80%, in the realms of "this is due to chance". [1]
Gives around a 5-6 month doubling relative to Opus 4 and a 6-7 month doubling relative to O3.
Imply evidence for the 50% - 80% capabilities systemically widening is weak.
[1] Note that this does provide evidence that Gemini 3 and GPT-5.2 will also have high p50 scores. Not because of capability jump per se but because of the distribution of tasks within METR benchmarks.
Could you share your model, if you haven't modeled it in an incognito chat? And don't forget to check out my comment below. If you are right, then the alternate Claude who succeeded at 15-second-long tasks would have a lower 50% time horizon.
P.S. I also asked the AIs a similar question. Grok 4 failed to answer, Gemini 3 Pro estimated the revised 80% horizon as 35-40 mins and the 50% horizon as 6-7 hrs. GPT-5's free version gave me this piece of slop, EDIT: Claude Opus 4.5 estimated the revised 80% horizons at 30-45 or 45-60 mins and 50% horizons as 2-3 hrs or 3-4 hrs.
Private workspace so I can’t share the session. But the approach is simple and doesn’t really require it to understand.
I think we’re coming at this from different angles: you’re doing a “white-box” critique (how specific task outcomes / curve fitting affect the METR horizon), whereas I’m doing a “black-box” consistency check: is the claimed p50 result consistent with what we see on other benchmarks that should correlate with capability?
The core model is:
Take Sonnet 4 → Sonnet 4.5 and compute the improvement rate (slope).
Assume Opus improves at the same rat... (read more)
An updated METR graph including Claude Opus 4.5 was just published 3 hours ago on X by METR (source):
Same graph but without the log (source):
Thread from METR on X (source):
The 80%-horizon has stayed essentially flat (27-32 mins) since GPT-5.1-Codex-Max's release but there's a big jump with huge error bars on the 50%-horizons.
I think Daniel Kokotajlo's recent shortform offers a useful framing here. He models progress as increasing either the intercept (baseline performance) or the slope (how well models convert time budget into performance). If progress comes mainly from rising intercepts, an exponential fit to horizon length could hold indefinitely. But if progress comes from increasing slope, the crossover point eventually shoots to infinity as AI slope approaches human slope.
The flat 80%-horizon while 50%-horizon climbs might be evidence for intercept-dominated progress.