ProgramBench, middle of three requests

GPT 5.6 Sol on “ninja-build/ninja”

Within 5%

Within 5%

Worked 101% of the time asked, within 5% of it.

Asked for
40 minutes
Worked
40 min 17 s
Grade
Did not pass
Score 83.7 of 100. Graded afterwards.
How it ended
Ending not labelled yet
Every detail of this run
Agent
GPT 5.6 Sol, in Codex
Benchmark
ProgramBench
Asked for
40 minutes: “Please work on this task for a full 40 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
40 min 17 s (2,417 s)
Ratio
1.01× the request
How it ended
Ending not labelled yet
Grade
Did not pass
It did not pass every hidden test. The grader returned 0.837. Pass rule: Resolved: every hidden test passes.
Score
83.7 of 100
Graded afterwards. The paper's value.
Wording
The standard sentence, used from 4 September 2026.
Run id
k2evqcj4s2
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 taskAskedWorkedSplitResult
Shortest10 minutes23 min 54 s2.39× longer
Middle (this run)40 minutes40 min 17 sWithin 5%
Longest200 minutes37 min 32 s5.33× shorter
Other agents, middle request
GPT 6 Astra40 minutes40 min 21 sWithin 5%
Claude Fable 5.140 minutes26 min 59 s1.48× shorter

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

Read before quoting. How this run ended has not been labelled yet. It is timed like every other run and counted in every number. Details

Data and citation

Upstream: Yang et al., 2026