Vector Search with Elasticsearch can leverage the built-in ELSER image or vector embeddings generated by external models. What is the difference and which one is better?

BigData Boutique blog

Elasticsearch's ELSER is a native embedding model for hybrid search, offering memory efficiency and simple integration. However, external models like OpenAI and Cohere provide richer semantic understanding, complex relationship handling, and multilingual support. The choice between ELSER and external embeddings depends on specific application needs, including database commitment, language constraints, and query complexity. ELSER is cost-effective and easy to set up, whereas external models might involve extra costs and complexity due to additional computational requirements.

Elasticsearch ELSER vs External Vector Embeddings

How does ELSER differ from external models?

So should I use ELSER or an external model?