Text Embedding 3 (Small) is 85% cheaper than Text Embedding 3 (Large) on blended 3:1 token pricing. Text Embedding 3 (Large) costs $0.13/1M in and $0.00/1M out, while Text Embedding 3 (Small) costs $0.02/1M in and $0.00/1M out.
Text Embedding 3 (Large) vs Text Embedding 3 (Small)
Compare verified input/output token rates, prompt caching discounts, latency tiers, and context limits between OpenAI and OpenAI.
| Metric / Feature | Text Embedding 3 (Large) (OpenAI) | Text Embedding 3 (Small) (OpenAI) |
|---|---|---|
| Input Price ($ / 1M Tokens) | $0.13 | $0.02 |
| Output Price ($ / 1M Tokens) | $0.00 | $0.00 |
| Prompt Caching Input Rate | Not Available | Not Available |
| Blended 3:1 Rate (Production) | $0.098 / 1M | $0.015 / 1M |
| Max Context Window | 8k | 8k |
| Multimodal Vision | ❌ Text Only | ❌ Text Only |
| MMLU Benchmark | N/A | N/A |
| Throughput & Latency | Fast | Ultra-Fast |
| Best For | Dense multilingual retrieval, legal/medical knowledge bases | Vector search, semantic clustering, RAG index creation |
When to Choose Text Embedding 3 (Large)
Choose Text Embedding 3 (Large) if your workload requires dense multilingual retrieval, legal/medical knowledge bases. Ideal for teams needing OpenAI's infrastructure and ecosystem tooling.
When to Choose Text Embedding 3 (Small)
Choose Text Embedding 3 (Small) if your primary objective is vector search, semantic clustering, rag index creation. Ideal for scaling high-throughput pipelines with OpenAI.
Want to compare Text Embedding 3 (Large) against another model?
Select another target model to view immediate pricing & latency trade-offs.