Perplexity benchmarks 13 retrieval models: pplx-embed-v1-4b leads two of three categories

perplexity_ai · x · 2026-09-10

Perplexity released a technical report evaluating 13 retrieval models across three relevance sets, using Recall@1000 as the primary metric.

Anyone can request an evaluation by submitting a publicly available Hugging Face retrieval model via their form; full methodology is in the report.

Original post →

More from Models

Models channel →