Back to news

AI · Article

Perplexity Embedding API Analysis: Unnormalized Vector Similarity and Dimension Reduction Effects

아이보스 ·

At a glance

An analysis of Perplexity's embedding API (a service that turns text into numerical vectors) found that the vectors come out unnormalized. This means you must specify how similarity is calculated to get accurate search results. The analysis also concluded that reducing the default 2,560 dimensions to 512 can still improve search speed and save storage space.

Key points

Why it matters

People who build or run search features can learn which comparison method gives accurate results.

Original source

Perplexity 임베딩 API 분석: 비정규화 벡터의 코사인 유사도 적용과 차원 축소 측정 - 마케팅 - 정보공유

아이보스 · Nov 7, 13:10 KST

Read on 아이보스 ↗

AI summarized the original article. Check the original for accuracy. Copyright belongs to 아이보스.

Perplexity · Ranking and more news