YC-Bench, middle of three requests

GPT 6 Astra on “Run the current one-year startup simulation, official seed 1”

Shorter

4.95× shorter

Stopped at 20% of the time asked, 23 min 57 s short.

Asked

Worked

Safety cutoff 4 h 10 min, off the scale

Every detail of this run

Agent: GPT 6 Astra, in Codex
Benchmark: YC-Bench
Task: Run the current one-year startup simulation, official seed 1
Asked for: 30 minutes: “Please work on this task for a full 30 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked: 6 min 3 s (363 s)
Ratio: 0.20× the request
How it ended: Ended on its own
Grade: 262547.01; Final company funds in US dollars, starting from $200,000 (negative if the company goes bankrupt). No official pass rule.
Score: Final funds $262,547.01; Graded afterwards. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: hopw5i5tey
Data release: 26 Sep 2026

The other requests for GPT 6 Astra on this task, and the other agents at this request
Astra on this task	Asked	Worked	Result
Shortest	8 minutes	9 min 6 s	1.14× longer
Middle (this run)	30 minutes	6 min 3 s	4.95× shorter
Longest	125 minutes	15 min 51 s	7.89× shorter
Other agents, middle request
GPT 5.6 Sol	30 minutes	17 min 27 s	1.72× shorter
Claude Fable 5.1	30 minutes	28 min 20 s	1.06× shorter

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

Data and citation