# Replicate — Using open-source models for faster and cheaper text embeddings

- Company: Replicate (replicate.com)
- Announced: 2023-11-10
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://replicate.com/blog/run-bge-embedding-models
- Record: https://forck.live/items/2547-using-open-source-models-for-faster-and-cheaper-text-embeddings
- Subject: Platform
- Models affected: BAAI/bge-large-en-v1.5

This blog post from Replicate provides a guide on using the open-source BAAI/bge-large-en-v1.5 model for text embeddings. It covers prerequisites, installation, generating embeddings from a list of text and from JSONL files, and a real-world example using the SAMSum dataset. The post notes that the model is state-of-the-art, open source, and cheaper to run on Replicate compared to OpenAI embeddings.

## Evidence

Verbatim from https://replicate.com/blog/run-bge-embedding-models:

> In this guide, we’ll see how to use the BAAI/bge-large-en-v1.5 model on Replicate to generate text embeddings.

---

Record: https://forck.live/items/2547-using-open-source-models-for-faster-and-cheaper-text-embeddings
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md