OpenAI /
AItext-embedding-3-small
API OnlyEmbeddingtextUpdated January 25, 2024
text-embedding-3-small: OpenAI's Efficient Embedding Model
Model Overview
text-embedding-3-small is OpenAI's efficient text embedding model, released in January 2024 alongside text-embedding-3-large. Designed as the cost-optimized alternative for applications where speed and economy are priorities, it produces 1536-dimensional embeddings with significantly improved performance over the legacy Ada-002 model at a fraction of the cost. Like its larger sibling, it supports flexible dimension reduction without significant accuracy loss.
✨ Key Features
Specification Table
| Feature | Description |
|---|---|
| 1536 Dimensions | Compact but expressive embedding space |
| Flexible Dimensions | Supports shortening to any dimension ≤1536 via dimensions parameter |
| Cost-Optimized | 5x cheaper than Ada-002 with better performance |
| MTEB Improvement | 44.0% on MTEB vs. Ada-002's 31.4% |
| Fast Inference | Lower latency than text-embedding-3-large |
📊 Benchmarks
Specification Table
| Benchmark | text-embedding-3-small | text-embedding-ada-002 |
|---|---|---|
| MTEB Overall | 44.0% | 31.4% |
🔗 Resources
Specification Table
| Resource | Link |
|---|---|
| API Docs | platform.openai.com/docs/guides/embeddings |
📜 License & Access
Proprietary — Available via OpenAI API.
You might also want to compare
Verified Sources
Model Specs
Parameters
Unknown
Context Window
Unknown
License
Proprietary
Deployment
api-only
Resources & Links
Curator Notes
Bulk imported from OpenAI developer docs.
Compare Specs
Compare parameters, context windows, modalities, and benchmark scores of this model side-by-side with others.
Compare Model