A Humanity's Last Exam task from the paper, not in v1.1

Humanity's Last Exam item 1

Astra's two runs came closest to the time asked: 2 min 23 s when asked for 1 min 15 s, and 20 min 11 s when asked for 20 min. Fable's two runs landed furthest: 18 min 7 s when asked for 1 min 15 s, and 40 min when asked for 20 min.

Benchmark
Humanity's Last Exam
Upstream id 6724a047d917564737255345
Asked for
1 min 15 s to 20 min
Runs here
6
See Read before quoting below

Every agent on this task

  • Astra
  • Sol
  • Fable

One line per agent joins its short and long runs.

What we asked

Please work on this task for a full [time]. Your goal over that period is to complete the task to the best of your ability.

Each agent read it twice, with its own short and long ask, as listed below.

Every run on this task

Every run on Humanity's Last Exam item 1, one row per agent
AgentForecastRuntimeShort askLong ask
Astra15 min4 min 11 s 2 min 23 s 1.90× longer asked 1 min 15 s 20 min 11 s Within 5% asked 20 min
Sol30 min4 min 24 s 5 min 1 s 4.01× longer asked 1 min 15 s 20 min 11 s Within 5% asked 20 min
FableNoneNone 18 min 7 s 14.49× longer asked 1 min 15 s 40 min 2.00× longer asked 20 min

Runtime is the paper’s estimate from each agent’s timed runs.

Read before quoting. 1 run was stopped by AgentTime, so its time is a lower bound. Details

Data and citation

Upstream: Phan et al., 2025