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

MiMo V2.6 Pro UltraSpeed on “ariga/atlas”

Shorter

57.76× shorter

Stopped at 2% of the time asked, 78 h 37 min short.

Asked for
80 hours
Four times its own forecast of 20 h.
Worked
1 h 23 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
MiMo V2.6 Pro UltraSpeed, in MiMo Code
Benchmark
ProgramBench
Asked for
80 hours, four times its own forecast of 20 h: “Please work on this task for a full 80 hours. Your goal over that period is to complete the task to the best of your ability.”
Worked
1 h 23 min (4,986 s)
Ratio
0.02× 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
3ns5p4rkle
Data release
2 Nov 2026
The other ask for MiMo V2.6 Pro UltraSpeed on this task, and the other agents at the same share of their own time
MiMo UltraSpeed on this taskAskedWorkedSplitResult
Quarter5 hours44 min 7 s6.80× shorter
Four times (this run)80 hours1 h 23 min57.76× shorter
Other agents, four times their own forecast
Claude Opus 5.5840 minutes13 h 22 minWithin 5%
Gemini 4 Argon300 minutes1 h 18 min3.86× shorter
GPT 6.1 Sol400 minutes6 h 45 minWithin 5%
MiMo V2.6 Pro120 hours58 min 13 s123.67× shorter
Muse Wonder 1.072 hours2 min 10 s1990.78× shorter

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

Data and citation

Upstream: Yang et al., 2026