Naget launches Trusted Web Search, claims 46% recall gain over Perplexity
HamedZamani · x · 2026-10-07
Search startup Naget introduced Trusted Web Search, treating source authority and provenance as first-class parts of search: it distinguishes institutional/expert sources from content farms, facts from opinions, and primary from secondary sources, letting people and agents search by source type they trust. The company claims 46% higher recall than Perplexity (GPT-6 Sol), Exa (xhigh) and Parallel (Ultra) on Perplexity's own WANDR dataset, plus 14% more perfect-recall queries on Total Recall QA (vendor-reported). It also offers an agent-optimized search API with bring-your-own-data or API-only options.
More from coding & agent
- DocETL in practice: a pipeline that extracts meds and side effects from transcripts — mdancho84 · 2026-10-07
- DocWrangler: Interactive playground for building AI document processing pipelines with DocETL — mdancho84 · 2026-10-07
- DocETL: A new Python library for agentic AI data processing and ETL — mdancho84 · 2026-10-07
- DocETL: a Python library for agentic data processing and ETL with AI — mdancho84 · 2026-10-07
- Josh Wills launches open-source data loading library — a year of commits as a fossil record of agent-assisted work — josh_wills · 2026-10-07
- AWS paper: structured agent communication (AECP) lifts multi-agent coding pass rate 28.2% — dair_ai · 2026-10-07