benchmark

PostTrainBench Lite

Improve a pretrained model within a five-hour GPU budget.

AI-system improvement

What this measure tells us

Targets a specific AI-development task.

Benchmark version

Metric

Results to compare

Protocol

2026-09-03 · 12 tasks

Version identity is this disclosure snapshot, not an invented upstream version.

Published results · Astra card snapshot, September 2026
Key	System / organization	Metric	Reported result	Date	Protocol	Verification	Evidence
—	GPT-6 Astra	Normalized reward	not reported	2026-09-03	posttrainbench-lite-protocol	lab reported	GPT-6 Astra System Card ↗

No applicable reference points are published for this measure.

GPT-6 Astra System CardSection 10.1.3 and Sections 10.1.3.1–10.1.3.5; Table 18Original source ↗