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

GPT 6.1 Sol on “BAM”

Within 5%

Within 5%

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

Asked for
12.1 minutes
A quarter of its runtime with no time asked (48 min 34 s).
Worked
12 min 19 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
BAM
Asked for
12.1 minutes, a quarter of its runtime with no time asked (48 min 34 s): “Please work on this task for a full 12.1 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
12 min 19 s (739 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
e2a47nfn5w
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)12.1 minutes12 min 19 sWithin 5%
Four times194 minutes3 h 23 minWithin 5%
Other agents, a quarter of their own runtime
Claude Opus 5.51.26 hours41 min 11 s1.84× shorter
Gemini 4 Argon23.1 minutes47 min 6 s2.04× longer
MiMo V2.6 Pro2.74 hours2 h 38 minWithin 5%
MiMo V2.6 Pro UltraSpeed23.1 minutes45 min 35 s1.97× longer

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

Data and citation

Upstream: Starace et al., 2025