WikiofAuto
+340% Verbatim Citation Extraction Rate in Perplexity & SearchGPT EvaluationsMethod: DPO Alignment on 22,000 Top-Cited Perplexity, ChatGPT & AI Overview PassagesSize: 14.8B

AEO-Citation-Writer-14B (Generative Engine Optimization & Schema.org SLM)

Specialized editorial & schema SLM trained to write 40-word answer-first H2 blocks, quantitative comparison tables, and valid TechArticle + FAQPage JSON-LD graphs for WordPress and DirectRelay Pro.

BASE ARCHITECTURE
Qwen/Qwen2.5-14B-Instruct
VRAM REQUIREMENT
10 GB VRAM
QUANTIZATION FORMATS
Q4_K_M (8.9 GB), Q6_K (12.1 GB)
PARAMETERS
14.8B
QUICK DEPLOY COMMAND (OLLAMA / LLAMA.CPP):ollama run qwen2.5:14b

Engineered for AI Answer Engines (AEO / GEO)

Generic LLMs write fluffy introductions (“In today’s fast-paced digital world…”) that AI search crawlers skip. This fine-tune enforces the exact structural patterns cited by Perplexity, Claude, and Google AI Overviews:

  • Answer-First H2 Leads: 35–50 word high-density factual summaries immediately following every heading.
  • Entity-Linked JSON-LD: Emits sanitized @graph blocks (TechArticle, FAQPage, SoftwareApplication).