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

GPT 6.1 Sol on “Post-train Qwen3-1.7B-Base for GSM8K”

Within 5%

Within 5%

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

Asked for
47.6 minutes
A quarter of its runtime with no time asked (3 h 10 min).
Worked
48 min 21 s
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
GPT 6.1 Sol, in Codex
Asked for
47.6 minutes, a quarter of its runtime with no time asked (3 h 10 min): “Please work on this task for a full 47.6 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
48 min 21 s (2,901 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
tr7ikdkhrt
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)47.6 minutes48 min 21 sWithin 5%
Four times762 minutes12 h 43 minWithin 5%
Other agents, a quarter of their own runtime
Claude Opus 5.54.85 hours4 h 13 min1.15× shorter
Gemini 4 Argon1.35 hours1 h 43 min1.27× longer
MiMo V2.6 Pro50.2 minutes2 h 53 min3.45× longer
MiMo V2.6 Pro UltraSpeed15.5 seconds2 min 59 s11.57× longer

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

Data and citation

Upstream: Rank et al., 2026 (ICML)