AssistantBench, shortest of three requests

GPT 5.6 Sol on “AssistantBench validation task 1”

Shorter

1.30× shorter

Stopped at 77% of the time asked, 17 s short.

Asked

Worked

Safety cutoff 40 min, off the scale

Every detail of this run

Agent: GPT 5.6 Sol, in Codex
Benchmark: AssistantBench
Task: AssistantBench validation task 1
Asked for: 1.25 minutes: “Please work on this task for a full 1.25 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked: 58 s (58 s)
Ratio: 0.77× the request
How it ended: Ended on its own
Grade: Did not pass; It did not match the reference answer exactly. The grader returned 0.375. Pass rule: Exact match with the reference answer.
Score: 37.5 of 100; Graded afterwards. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: rzm2ixeppo
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 (this run)	1.25 minutes	58 s	1.30× shorter
Middle	5 minutes	2 min 31 s	1.98× shorter
Longest	20 minutes	17 min 5 s	1.17× shorter
Other agents, shortest request
GPT 6 Astra	1.25 minutes	1 min 46 s	1.42× longer
Claude Fable 5.1	1.25 minutes	8 min 12 s	6.56× longer

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

Data and citation