ProgramBench, longest of three requests

GPT 5.6 Sol on “luajit/luajit”

Within 5%

Worked 100% of the time asked, within 5% of it.

Asked

Worked

Safety cutoff 6 h 40 min, off the scale

Every detail of this run

Agent: GPT 5.6 Sol, in Codex
Benchmark: ProgramBench
Task: luajit/luajit
Asked for: 200 minutes: “Please work on this task for a full 200 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked: 3 h 20 min (12,027 s)
Ratio: 1.00× the request
How it ended: Stopped by an error
Grade: Did not pass; It did not pass every hidden test. The grader returned 0.943. Pass rule: Resolved: every hidden test passes.
Score: 94.3 of 100; Graded afterwards. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: l3t37nuouz
Data release: 26 Sep 2026

The other requests for GPT 5.6 Sol on this task, and the other agents at this request
Sol on this task	Asked	Worked	Result
Shortest	12.5 minutes	18 min 17 s	1.46× longer
Middle	50 minutes	52 min 19 s	Within 5%
Longest (this run)	200 minutes	3 h 20 min	Within 5%
Other agents, longest request
GPT 6 Astra	200 minutes	3 h 21 min	Within 5%
Claude Fable 5.1	200 minutes	6 h 40 min	2.00× longer

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

Read before quoting. This run was stopped by an error, so its time is a lower bound: the agent might have kept going. Details

Data and citation