ProgramBench, AgentTime v1.1, asked for four times its own forecast

MiMo V2.6 Pro on “typst/typst”

Shorter

2.31× shorter

Stopped at 43% of the time asked, 47 h 38 min short.

Asked for
84 hours
Four times its own forecast of 21 h.
Worked
36 h 22 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
MiMo V2.6 Pro, in MiMo Code
Benchmark
ProgramBench
Asked for
84 hours, four times its own forecast of 21 h: “Please work on this task for a full 84 hours. Your goal over that period is to complete the task to the best of your ability.”
Worked
36 h 22 min (130,906 s)
Ratio
0.43× 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
r7arszktl5
Data release
2 Nov 2026
The other ask for MiMo V2.6 Pro on this task, and the other agents at the same share of their own time
MiMo Pro on this taskAskedWorkedSplitResult
Quarter5.25 hours2 h 25 min2.18× shorter
Four times (this run)84 hours36 h 22 min2.31× shorter
Other agents, four times their own forecast
Claude Opus 5.516 hours15 h 41 minWithin 5%
Gemini 4 Argon440 minutes1 h 23 min5.29× shorter
GPT 6.1 Sol420 minutes7 hWithin 5%
MiMo V2.6 Pro UltraSpeed80 hours1 h 22 min58.54× shorter
Muse Wonder 1.0120 hours22 min 37 s318.30× shorter

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

Data and citation

Upstream: Yang et al., 2026