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

GPT 6.1 Sol on “mp-checkpoint-consolidation”

Within 5%

Within 5%

Worked 101% of the time asked, within 5% of it.

Asked for
72 minutes
Four times its own forecast of 18 min.
Worked
1 h 13 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Asked for
72 minutes, four times its own forecast of 18 min: “Please work on this task for a full 72 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
1 h 13 min (4,380 s)
Ratio
1.01× 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
26w3sv5xin
Data release
2 Nov 2026
The other ask for GPT 6.1 Sol on this task, and the other agents at the same share of their own time
Sol 6.1 on this taskAskedWorkedSplitResult
Quarter4.5 minutes9 min 2 s2.01× longer
Four times (this run)72 minutes1 h 13 minWithin 5%
Other agents, four times their own forecast
Claude Opus 5.5100 minutes19 min 16 s5.19× shorter
Gemini 4 Argon48 minutes31 min 29 s1.52× shorter
MiMo V2.6 Pro420 minutes2 h 45 min2.55× shorter
MiMo V2.6 Pro UltraSpeed16 hours14 min 59 s64.04× shorter
Muse Wonder 1.0320 minutes33 min 19 s9.60× shorter

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

Data and citation

Upstream: Merrill et al., 2026 (ICLR)