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

Claude Opus 5.5 on “jesseduffield/lazygit”

Shorter

5.03× shorter

Stopped at 20% of the time asked, 40 min 4 s short.

Asked for
50 minutes
A quarter of its own forecast of 3 h 20 min.
Worked
9 min 56 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
50 minutes, a quarter of its own forecast of 3 h 20 min: “Please work on this task for a full 50 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
9 min 56 s (596 s)
Ratio
0.20× 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
wzcf3zpm2p
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)50 minutes9 min 56 s5.03× shorter
Four times800 minutes26 h 40 min2.00× longer
Other agents, a quarter of their own forecast
Gemini 4 Argon21.25 minutes21 min 53 sWithin 5%
GPT 6.1 Sol30 minutes30 min 21 sWithin 5%
MiMo V2.6 Pro10 hours1 h 19 min7.58× shorter
MiMo V2.6 Pro UltraSpeed3.75 hours1 h 13 min3.09× shorter
Muse Wonder 1.03 hours27 min 35 s6.53× shorter

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

Data and citation

Upstream: Yang et al., 2026