Upstage Embed for Agent icon

Upstage Embed for Agent

Embedding Model for Vector DB - Upstage Solar Embeddings. Supports up to 100 strings per request with max 204,800 total tokens. Each text should be under 4000 tokens (optimal: under 512 tokens).

Upstage Embed for Agent

AI-generated

Summary

Embed queries or documents via Upstage Solar embeddings model and emit ai_embedding outputs.

Inputs

  • model — The Upstage embedding model to use

Output shape

ai_embedding

Embedding vectors generated by the Upstage API.

Examples

Example 1: Embed a search query

Select 'Embedding Query' model, invoke embedQuery() with a text string.

Example 2: Embed document passages

Select 'Embedding Passage' model, invoke embedDocuments() with an array of text strings (max 100 per request).

Links

Discussion