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

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

Shorter

31.36× shorter

Stopped at 3% of the time asked, 11 h 37 min short.

Asked for
720 minutes
Four times its own forecast of 3 h.
Worked
22 min 58 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
MiMo V2.6 Pro, in MiMo Code
Asked for
720 minutes, four times its own forecast of 3 h: “Please work on this task for a full 720 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
22 min 58 s (1,378 s)
Ratio
0.03× 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
gzo3ffzklu
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
Quarter45 minutes22 min 14 s2.02× shorter
Four times (this run)720 minutes22 min 58 s31.36× 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 Pro UltraSpeed380 minutes8 min 1 s47.38× 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)