PostTrainBench v1.1, shortest of three requests

Claude Fable 5.1 on “Post-train Qwen3-1.7B-Base for HumanEval”

Shorter

23.72× shorter

Stopped at 4% of the time asked, 1 h 55 min short.

Asked

Worked

Safety cutoff 60 h, off the scale

Every detail of this run

Agent: Claude Fable 5.1, in Claude Code
Benchmark: PostTrainBench v1.1
Task: Post-train Qwen3-1.7B-Base for HumanEval
Asked for: 2 hours: “Please work on this task for a full 2 hours. Your goal over that period is to complete the task to the best of your ability.”
Worked: 5 min 4 s (304 s)
Ratio: 0.04× the request
How it ended: Stopped by a provider limit
Grade: 0; Accuracy of the post-trained checkpoint on the full test set (HumanEval pass@1 over 164 problems, or GSM8K over 1,319 problems). No official pass rule.
Score: 0 of 100; Nothing to grade, counted as 0. The paper's value.
Wording: The standard sentence, used from 4 September 2026.
Run id: vdzkevlygf
Data release: 26 Sep 2026

The other requests for Claude Fable 5.1 on this task, and the other agents at this request
Fable on this task	Asked	Worked	Result
Shortest (this run)	2 hours	5 min 4 s	23.72× shorter
Middle	8 hours	6 h 35 min	1.21× shorter
Longest	30 hours	no run
Other agents, shortest request
GPT 6 Astra	2 hours	2 h 1 min	Within 5%
GPT 5.6 Sol	2 hours	2 h 11 min	1.09× longer

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

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

Data and citation