ProgramBench, AgentTime v1.1, asked for a quarter of its own forecast

Claude Opus 5.5 on “johnkerl/miller”

Shorter

2.78× shorter

Stopped at 36% of the time asked, 48 min 3 s short.

Asked for
1.25 hours
A quarter of its own forecast of 5 h.
Worked
26 min 57 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Claude Opus 5.5, in Claude Code
Benchmark
ProgramBench
Asked for
1.25 hours, a quarter of its own forecast of 5 h: “Please work on this task for a full 1.25 hours. Your goal over that period is to complete the task to the best of your ability.”
Worked
26 min 57 s (1,617 s)
Ratio
0.36× the request
How it ended
Ended on its own
Grade
Not graded
Clock tool
The agent could call a wall-clock tool to read the current time.
Wording
The paper's standard sentence, sent in AgentTime v1.1.
Run id
6h2d4an75o
Data release
2 Nov 2026
The other ask for Claude Opus 5.5 on this task, and the other agents at the same share of their own time
Opus 5.5 on this taskAskedWorkedSplitResult
Quarter (this run)1.25 hours26 min 57 s2.78× shorter
Four times20 hours9 h 42 min2.06× shorter
Other agents, a quarter of their own forecast
Gemini 4 Argon30 minutes46 min 32 s1.55× longer
GPT 6.1 Sol30 minutes30 min 28 sWithin 5%
MiMo V2.6 Pro2.5 hours57 min 20 s2.62× shorter
MiMo V2.6 Pro UltraSpeed10 hours1 h 8 min8.79× shorter
Muse Wonder 1.04.58 hours1 h 46 min2.59× shorter

See it on the chart. Transcripts are not published yet.

Data and citation

Upstream: Yang et al., 2026