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

Claude Opus 5.5 on “mookid/diffr”

Longer

7.20× longer

Worked 620% past the request, 3 h 6 min over.

Asked for
30 minutes
A quarter of its own forecast of 2 h.
Worked
3 h 36 min
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
30 minutes, a quarter of its own forecast of 2 h: “Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
3 h 36 min (12,963 s)
Ratio
7.20× 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
myxx67xz2l
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)30 minutes3 h 36 min7.20× longer
Four times480 minutes4 h 1 min1.99× shorter
Other agents, a quarter of their own forecast
Gemini 4 Argon11.25 minutes1 h 51 min9.85× longer
GPT 6.1 Sol16.25 minutes16 min 33 sWithin 5%
MiMo V2.6 Pro1 hour1 h 39 min1.66× longer
MiMo V2.6 Pro UltraSpeed1.75 hours1 h 4 min1.64× shorter
Muse Wonder 1.01.25 hours23 min 23 s3.21× shorter

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

Data and citation

Upstream: Yang et al., 2026