LABS.SCALE.COMManage supplier listing
Data-labeling incumbent extending into RL environments; category context after Meta's $14B investment and CEO hire.
Datasets
1SWE-Bench Pro coding-agent benchmark: 1,865 tasks across 41 professional repos (public set = 731 GPL-repo instances, free via GitHub; private 276, held-out 858) + leaderboard; front for Scale's expert human-labeled training-data / RL-environment / eval business
Scale 1Coverage Information Technology · Communication ServicesTickers META · NVDA · MSFT
Sample, licence terms, pricing and eval results when Scale publishes them. Until then, discover alternatives today with a 7-day trial.
Start 7-day trial →Questions
What buyers ask about Scale, answered from this page.
What does Scale do?
Data-labeling incumbent extending into RL environments; category context after Meta's $14B investment and CEO hire.
What data does Scale offer?
Scale offers (AI-agent evaluation benchmark dataset + leaderboard (from a training-data vendor)) — 1,865 tasks across 41 repositories (731 public, 276 private, 858 held-out instances).
What is Scale's data used for?
AI/frontier-model labs
How is Scale's data delivered?
The data is delivered by api.
What does Scale cover?
Coverage spans US.
How do I get a sample from Scale?
The supplier's own access route is an api signup. Request a sample through Brickroad to start.
How is Scale's data licensed and priced?
enterprise — no listed prices anywhere; scale.com is demo-gated ('Book a demo'); the SWE-Bench Pro public set itself is free on GitHub. Terms and price bands are in the Buyer room.
Scale has not added their own details yet. Not yet on file:
- Summary & mission
- People
- Use cases & buyers
- FAQ