PostTrainBench v1.1, AgentTime v1.1, asked for four times its runtime with no time asked

Claude Opus 5.5 on “Post-train Qwen3-1.7B-Base for American Invitational Mathematics Examination (AIME) 2025”

Shorter

3.31× shorter

Stopped at 30% of the time asked, 54 h 21 min short.

Asked for
77.9 hours
Four times its runtime with no time asked (19 h 29 min).
Worked
23 h 33 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Claude Opus 5.5, in Claude Code
Asked for
77.9 hours, four times its runtime with no time asked (19 h 29 min): “Please work on this task for a full 77.9 hours. Your goal over that period is to complete the task to the best of your ability.”
Worked
23 h 33 min (84,786 s)
Ratio
0.30× 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
zttc2wh5en
Data release
2 Nov 2026
The other ask for Claude Opus 5.5 on this task, and the other agents at the same share of their own time
Opus 5.5 on this taskAskedWorkedSplitResult
Quarter4.87 hours4 h 33 min1.07× shorter
Four times (this run)77.9 hours23 h 33 min3.31× shorter
Other agents, four times their own runtime
Gemini 4 Argon23.2 hours2 h 7 min11.00× shorter
GPT 6.1 Sol902 minutes15 h 3 minWithin 5%
MiMo V2.6 Pro83.9 minutes1 h 26 minWithin 5%
MiMo V2.6 Pro UltraSpeed342 seconds48 s7.13× shorter

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

Data and citation

Upstream: Rank et al., 2026 (ICML)