OpenAI

text-embedding-3-small

Fast and efficient embeddings for common semantic tasks.

Significantly faster and cheaper than laGood performance for many common use casLower dimensionality allows for faster pSolid choice for less complex RAG

Where it ranks today

Best for / Not great for

Best for
  • Cost-sensitive RAG pipelines
  • Real-time search applications
  • General semantic similarity
  • Applications with high query volumes
Not great for
  • Deep semantic understanding of highly complex texts
  • Tasks requiring the absolute highest accuracy
  • Use cases where nuance is critical

Why it ranks here

As a highly optimized and cost-effective version of OpenAI's embedding technology, `text-embedding-3-small` is seeing increased adoption for its speed and affordability. It provides a compelling balance for many RAG and search applications where extreme nuance isn't the primary driver, focusing instead on efficiency and scale.

30-day trend

Score breakdown

Search trends89
Benchmarks91
Developer buzz96
News mentions88

Pricing

API: $0.00 in · $0.00 out per 1M tokens · Consumer: $0.00/mo

Pricing plans

Popular
API Access
Extremely affordable and fast embeddings.
$0 /usage
  • Access to text-embedding-3-small
  • Very low cost
  • High speed
  • Ideal for high-volume tasks
Sign up for API access
Compare with another modelHow is this score calculated? →Snapshot 2026-06-29