BGE-M3 API

BGE-M3 is an embedding model by BAAI, available on the Venice API as text-embedding-bge-m3. It runs privately, with zero data retention.

BGE-M3 API pricing

Model ID	Variant	Privacy	Price
`text-embedding-bge-m3`	Standard	Private	$0.15 per 1M tokens

BGE-M3 specifications

Spec	Value
Provider	BAAI
Released	Mar 14, 2025
Privacy	Private
Open weights	Yes
Dimensions	1,024
Max input	8K tokens

How to use the BGE-M3 API

Send requests to POST https://api.venice.ai/api/v1/embeddings with "model": "text-embedding-bge-m3" and your API key.

curl https://api.venice.ai/api/v1/embeddings \
  -H "Authorization: Bearer $VENICE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-bge-m3",
    "input": "Private AI for everyone."
  }'

BGE-M3 API FAQ

How much does the BGE-M3 API cost?

$0.15 per 1M input tokens. Prices are in USD and can be paid in DIEM at parity.

What is the BGE-M3 model ID?

Use text-embedding-bge-m3 as the model parameter.

Is the BGE-M3 API private?

BGE-M3 is private: requests run on infrastructure Venice controls with zero data retention, and prompts and outputs are never stored or used for training.

How many dimensions do BGE-M3 embeddings have?

1,024 dimensions, with up to 8K input tokens per item.

Which endpoint does the BGE-M3 API use?

Call POST /embeddings.

Related models

Text Embedding 3 Large API: $0.16 per 1M tokens
Gemini Embedding 2 Preview API: $0.25 per 1M tokens
Text Embedding 3 Small API: $0.03 per 1M tokens
Multilingual E5 Large Instruct API: $0.01 per 1M tokens

Explore

By modality