French Corpus LLM Sample: 500 Stratified Documents from Finance and Regulatory Open Data

500 stratified documents form a public sample drawn from the French Premium Web Corpus v1.4.0. FINALEADS LLC built this compliance-ready training dataset from 2.66 billion tokens of French finance, regulatory, and economic open data. The dataset was last updated on 2026-05 -19.

Use Cases

• Training French language models based on the corpus of finance and regulatory text.
• Developing compliance and regtech tools based on the documented, pseudonymized data.
• Analyzing economic and financial language patterns based on the stratified sample of open data.

Strengths

• Sample is drawn from a larger corpus of 2.66 billion tokens.
• Data is processed to be audit-trailed, pseudonymized, and documented for AI Act Article 10 compliance.
• Documents are stratified, suggesting a methodical sampling approach.

Limitations

• Row count for the full dataset is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

FINALEADS LLC, using French finance, regulatory, and economic open data.

Collection Method

Stratified sampling from the French Premium Web Corpus v1.4.0.

Time Range

null

Freshness

Last updated 2026-05-19 20:14:08; freshness should be verified.

Geography

France