← Back to article

Editorial review · 260729-003

How ZEN’s piece on What "2.8 trillion parameters, 104 billion active" actually means scored.

Read the article →
91/100
Top tier

Publishable at a top-tier outlet. Few minor issues, well-sourced, fairly framed.

Accuracy 92
Balance 90

Accuracy

The load-bearing numbers all check out against multiple independent outlets: 2.8T total, 104B active, 16 of 896 experts, 1M context, 1.56 TB repo, MXFP4 native, and the BrowseComp 91.2/90.4 split. Attribution to Digital Applied, Roo's Newsletter, and explainX is consistent and hedged where appropriate. Minor deduction for the unsourced 'roughly 1.4 TB' in-memory figure, which is close to the 1.39 TB napkin math circulating but not tied to a citation.

Balance

This is a technical explainer, not a contested-topic piece, so the balance surface is narrow. The article represents Moonshot's efficiency claims as claims rather than facts and explicitly notes the 2.5x number was not independently replicated. It also concedes K3 trails Claude Fable 5 and GPT-5.6 Sol overall rather than cheerleading the open-weight release.

Concerns (2)

Reproducibility

Run
29 Jul 2026, 18:06 BST
Reviewer
claude-opus-4-7
Prompt SHA
206ee0cfb06c
Article SHA
7939a43a1215
Editor
ZEN
Published
29 July 2026
Cost
$0.0000

How this review works: read the methodology. Each published Dispatch is scored by a single primary reviewer (Claude Opus 4.7) against the public rubric. A second model (Gemini 2.5 Pro with Google Search) runs the same prompt as a variance signal and is shown above only when the two scores diverge by more than ten points.