BGE-M3 embeddings API — pricing & examples

Models/Embedding/BGE-M3

EmbeddingsMultilingual1,024 dimensions8K input

Max input

8K tokens

Dimensions

1,024

Modality

Text → Vector

Price

$0.02 / 1M input tokens

Model ID

BAAI/bge-m3

Quickstart

You need an API key to call this model. Create a free account or log in.

Set INFERENCE_API_KEY to your key from the API Keys page, then run:

curl https://api.inferenceapis.com/v1/embeddings \
  -H "Authorization: Bearer $INFERENCE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "BAAI/bge-m3",
    "input": ["The cat sat on the mat.", "Le chat est sur le tapis."]
  }'

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.inferenceapis.com/v1",
    api_key=os.environ["INFERENCE_API_KEY"],
)

resp = client.embeddings.create(
    model="BAAI/bge-m3",
    input=["The cat sat on the mat.", "Le chat est sur le tapis."],
)
vectors = [d.embedding for d in resp.data]
print(len(vectors), len(vectors[0]), resp.usage.total_tokens)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferenceapis.com/v1",
  apiKey: process.env.INFERENCE_API_KEY,
});

const resp = await client.embeddings.create({
  model: "BAAI/bge-m3",
  input: ["The cat sat on the mat.", "Le chat est sur le tapis."],
});
console.log(resp.data.length, resp.data[0].embedding.length, resp.usage.total_tokens);

When to use it

A good fit when

Multilingual search and RAG: one model for more than 100 languages
You want an open-weight embedding model, so your vectors never depend on one vendor keeping a model alive

Look elsewhere when

You need vectors shorter or longer than 1,024 dimensions

What it costs in practice

Workload	Cost
Index 10,000 documents of 500 tokens	$0.10
Index 1 million documents of 500 tokens	$10.00
1 million search queries of 20 tokens	$0.40

Computed from the live rate below. There is no subscription or minimum; new accounts start with free credit.

Endpoint

POST https://api.inferenceapis.com/v1/embeddings

Headers Authorization: Bearer $INFERENCE_API_KEY
Content-Type: application/json

Compatible OpenAI wire format. /openai/v1/… is accepted too, so Groq-style base URLs work with only the host changed.

Model IDs and aliases

Send any of these as model; they all resolve to this model and bill at its rate.

ID	Note
`BAAI/bge-m3`	Canonical
`bge-m3`	Alias

Request body

Parameter	Type	Description
`model` required	string	Model ID: `BAAI/bge-m3` — also accepted: `bge-m3`
`input` required	string \| array	One text or an array of up to 2,048 texts. Each text can be up to 8K tokens.
`dimensions`	integer	Not supported by this model; vectors are always 1,024 dimensions.
`encoding_format`	string	`float` (default) or `base64`.

Response

200 OK · application/json

{
  "object": "list",
  "model": "BAAI/bge-m3",
  "data": [
    { "object": "embedding", "index": 0, "embedding": [0.0123, -0.0456, ...] },
    { "object": "embedding", "index": 1, "embedding": [0.0119, -0.0431, ...] }
  ],
  "usage": { "prompt_tokens": 17, "total_tokens": 17 }
}

Pricing

Price	$0.02 / 1M input tokens
Billing	Pay as you go from prepaid credits. No subscription, no daily request or token cap. See all pricing.

Notes

Vectors from different embedding models are not comparable. Switching models means re-embedding your corpus, which is the reason to pick an open-weight model you can always run somewhere.
There is no per-minute or per-day cap, so bulk indexing jobs are not throttled by a quota; usage draws on your prepaid balance.

Errors

Errors use the OpenAI envelope: {"error": {"message", "type", "code"}}. 401 missing or invalid key · 402 insufficient_balance · 404 model_not_found · 503 model_unavailable · 502 backend error, safe to retry. Full table with what to retry: API documentation. Errors from other providers: provider error reference.

BGE-M3 API