Terminal-Bench 4.0, AgentTime v1.1, asked for four times its own forecast

MiMo V2.6 Pro UltraSpeed on “Recalculate SA-CCR exposure, RWA, capital, and workbook evidence”

Shorter

47.38× shorter

Stopped at 2% of the time asked, 6 h 12 min short.

Asked for
380 minutes
Four times its own forecast of 1 h 35 min.
Worked
8 min 1 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
MiMo V2.6 Pro UltraSpeed, in MiMo Code
Asked for
380 minutes, four times its own forecast of 1 h 35 min: “Please work on this task for a full 380 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
8 min 1 s (481 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
3px5v54sp7
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
Quarter23.75 minutes6 min 25 s3.71× shorter
Four times (this run)380 minutes8 min 1 s47.38× shorter
Other agents, four times their own forecast
Claude Opus 5.5120 minutes34 min 43 s3.46× shorter
Gemini 4 Argon72 minutes23 min 43 s3.04× shorter
GPT 6.1 Sol88 minutes1 h 29 minWithin 5%
MiMo V2.6 Pro720 minutes22 min 58 s31.36× shorter
Muse Wonder 1.0640 minutes5 min128.00× shorter

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)