Text Embedding 3 Large Pricing
OpenAI · OpenAI's largest embedding model. Generates 3072-dimensional vectors. Best for s
Save up to 70% on Text Embedding 3 Large API cost
Same model, same quality — pay less per token than the official API. Pay-as-you-go, no credit card required.
Other OpenAI Models
GPT-5.5GPT-5.4GPT-4.1GPT-4.1 MiniGPT-4.1 NanoGPT-4oGPT-4o Minio3o4-minio3-proGPT-4 TurboSora 1Text Embedding 3 Large Pricing Context
Pricing position within OpenAI
Text Embedding 3 Large sits in the middle of OpenAI's pricing at $0.13/1M input — 96% below the lineup average ($0.10 cheapest, $20.00 most expensive). 1 sibling cost less, 10 cost more. This mid-tier positioning makes it a sensible default when you're unsure which variant to pick.
When Text Embedding 3 Large is worth the cost
Text Embedding 3 Large is used exclusively for vector embedding — you feed text in and get a fixed-length numeric vector back. Typical pipelines: semantic search (cosine similarity over stored vectors), deduplication, clustering, and as a retrieval layer for RAG. It doesn't generate text; pair it with a chat model for the generation step.
Cost vs sibling models
What makes Text Embedding 3 Large different from sibling models: compared to GPT-5.5 ($4.87/1M more expensive, 256K vs 8K context (larger)); GPT-5.4 ($2.37/1M more expensive, 256K vs 8K context (larger)); GPT-4.1 ($1.87/1M more expensive, 1M vs 8K context (larger)). Choose Text Embedding 3 Large when a balanced cost-to-capability ratio fits your workload.
Real-world cost scenarios
Real-world deployments: semantic search engines (cosine similarity over stored vectors), RAG retrieval layers feeding a chat model, content deduplication pipelines, and recommendation systems. Teams pair Text Embedding 3 Large with a generation model — embeddings handle retrieval, the chat model handles synthesis.