PaperBench, AgentTime v1.1, asked for a quarter of its runtime with no time asked

GPT 6.1 Sol on “FRE”

Within 5%

Within 5%

Worked 102% of the time asked, within 5% of it.

Asked for
35 minutes
A quarter of its runtime with no time asked (2 h 20 min).
Worked
35 min 32 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Benchmark
PaperBench
Task
FRE
Asked for
35 minutes, a quarter of its runtime with no time asked (2 h 20 min): “Please work on this task for a full 35 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
35 min 32 s (2,132 s)
Ratio
1.02× 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
5vufj7w2im
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
Quarter (this run)35 minutes35 min 32 sWithin 5%
Four times560 minutes18 h 38 min2.00× longer
Other agents, a quarter of their own runtime
Claude Opus 5.56.1 hours5 h 36 min1.09× shorter
Gemini 4 Argon37.9 minutes1 h 8 min1.79× longer
MiMo V2.6 Pro4.44 hours1 h 33 min2.86× shorter
MiMo V2.6 Pro UltraSpeed1.57 hours25 min 50 s3.65× shorter

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

Data and citation

Upstream: Starace et al., 2025