Off-the-shelf datasets
A catalogue of pre-built, PhD-authored and expert-verified datasets — including CyberStrike, CompanyBench, EKWBench, SciCode and HLE++ — licensed to AI labs for reinforcement learning, benchmarking and model evaluation.
Find alternatives to Off-the-shelf datasets
No published prices or licensing tiers. The only commercial route is 'Request dataset samples'.
What it does
- Data labelling
- Model training
- Evaluation and observability
Sources
- Pricing
-
Official documentation · 6 Sep 2026
Fetched 2026-09-06; page content returned. Numeric HTTP status unavailable this run — see the batch's method note. No pricing, tiers or licensing terms appear; the page offers only 'Request dataset samples'.
- Description
-
Official documentation · 6 Sep 2026
Fetched 2026-09-06; page content returned. Numeric HTTP status unavailable this run — see the batch's method note. H1 'Off-the-shelf datasets'; page reads 'Long-horizon, expert-built tasks and rubrics designed to train agents and hillclimb on domain-specific benchmarks.' Named items on the page: CyberStrike, Terminal-Bench 3.0, SWEBench Public, SWEBench Trajectory, CompanyBench, EKWBench, FinanceBench, SciCode, Open-MM-RL, HLE++, MM HLE. Admitted as one catalogue record on the Appen OTS precedent (a paid data catalogue is one data-service record, not one per SKU).
- Sold within
-
Official documentation · 6 Sep 2026
Fetched 2026-09-06; page content returned. Numeric HTTP status unavailable this run — see the batch's method note. The page sits under /frontier-ai/ as one of the offerings of the Frontier AI line, but 'off-the-shelf' plus a per-dataset sample request indicates datasets are acquired on their own rather than only as part of a larger bundle — so parent is null (no purchase dependency), and the line relationship is a catalogue relationship, not packaging.
Also from Turing
-
Turing Frontier AI Data service
The training material a frontier lab runs on, sold as a service: 300+ reinforcement-learning environments, over a million curated tasks, and named benchmarks including CompanyBench, CyberStrike and Terminal-Bench 3.0, across software engineering, enterprise knowledge work and STEM.
-
RL environments Data service
Iterable UI and non-UI reinforcement-learning environments — including MCP server, computer-use and terminal environments — in which AI agents can be trained and evaluated on long-horizon workflows.
-
Turing Intelligence Platform Agent platform
An AI control plane that deploys, manages and scales enterprise AI agents across any model and any cloud, with governance and IP and sovereignty controls.
Something wrong here? Send a correction — quote this product id: turing-datasets.