ActiveREQ-000015

Seeking Large-Scale Technical Document Datasets

We are looking to acquire rare, specialized, and technical document datasets for AI pretraining.

Requirement brief

Specific requirements

Document type: Preferably long-form, technical, and specialized documents.
Formats: PDF, DOC/DOCX, or other available formats.
Accessibility: Data should not be easily scrapeable or readily available on the public Internet.
Sample: An initial sample of 300–400 documents is required. The sample must be representative of the overall dataset.
Dataset estimate: Total document count and estimated token volume should be provided.
Information required — 1. Document types and domains.
Information required — 2. Total document count and estimated token volume.
Information required — 3. Available formats.
Information required — 4. Data ownership and licensing/usage rights.
Information required — 5. Availability of a representative 300–400 document sample.
Information required — 6. Pricing and delivery timeline.

Coverage

Review the supplier process and prepare your capabilities.