Structured, AI-processed global mining technical-report corpus

Name: Structured, AI-processed global mining technical-report corpus: 7,367 NI 43-101 / SK-1300 / JORC-competent-person reports spanning 3,792 companies, 136 countries and 142 commodities, indexed and chunked with page-level citations, plus a company-project-commodity-person knowledge graph and model-derived project attributes (capex, production rate, dilution, recovery, infrastructure, emissions and water risks, with page references such as [p. 425]).
Creator: MineGPT

MINEGPTManage supplier listing

Structured, AI-processed global mining technical-report corpus: 7,367 NI 43-101 / SK-1300 / JORC-competent-person reports spanning 3,792 companies, 136 countries and 142 commodities, indexed and chunked with page-level citations, plus a company-project-commodity-person knowledge graph and model-derived project attributes (capex, production rate, dilution, recovery, infrastructure, emissions and water risks, with page references such as [p. 425]).

Open buyer room →

Sample

Brickroad customers

Sample this dataset before you buy it.

Your sourcing agent asks MineGPT and files the sample in your catalog — private to you.

Start 7-day trialNothing is charged until the trial ends.

Coverage

Universe, instruments and categories.

Industry: Diversified Metals & MiningApplication Software
Instruments: equitiescommodities

Dataset card

Type: not stated
Format: Multimodal — source PDFs (technical reports, downloadable in-app) plus a derived graph/tabular layer (entities and relations for the General Knowledge Graph) and text embeddings/extracted tables used for retrieval with page-level provenance
Volume: Stated: 7,367 technical reports, 3,792 companies, 136 countries, 142 commodities. At several hundred pages per technical report this implies a corpus on the order of 2-4 million document pages, plus the derived entity/relation graph over those reports.
Users: N/A for consumers — paid-subscription B2B platform ('Paid plans only' gating the knowledge graph); no subscriber count published. Est. low hundreds of professional seats inferred from a one-person organisation operating an AI-native product.
History: 25 years
Update frequency: not stated
Growth: Active — live coverage counters, 2025-dated feasibility studies already ingested (e.g. a 2025 NI 43-101 for a named project), a demo video and a paid-plan gating funnel; site built on a rapid-iteration front-end stack.
Launched: ~2024-2025 (modern AI-native product with LLM chat as the core interface, a rapid-build front-end stack, and a single-person organisation in company records).
Delivery: not stated
Entity mapping: not stated
Sample: not stated
Point-in-time: not stated
Licence: not stated

supplier index read from their site