Text embedding model

BGE-M3 API

BAAI open multilingual text embeddings, 1024 dimensions, for retrieval across 100+ languages.

BGE-M3 is a text embedding model from BAAI, available on OttoPort's OpenAI-compatible API as the model ID bge-m3 at $0.011 per 1M input tokens as of 2026-10-14.

Provider

BAAI

Price

$0.011 per 1M input tokens

Model ID

bge-m3

Type

text embedding model

Semantic searchRetrieval (RAG)Clustering and dedup

Pricing

BGE-M3 API pricing

What OttoPort bills for BGE-M3, and what that works out to per unit. Rates are read from the same catalog the gateway meters against, and were current on 2026-10-14.

Per 1M tokens	$0.011	input text; the vectors are not billed
10,000 passages	$0.055	of about 500 tokens each, embedded once
Tokens per $1	90,909,090

No subscription and no minimum — OttoPort bills prepaid credits against usage. See the full rate card.

Quickstart

Call BGE-M3 in three lines

Send "model": "bge-m3" to POST https://ottoport.ai/api/v1/embeddings with an OttoPort key. No BAAI account, no waitlist, no separate SDK.

curl https://ottoport.ai/api/v1/embeddings \
  -H "Authorization: Bearer $OTTOPORT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "bge-m3",
    "input": ["How do refunds work?", "Refunds are issued within 5 days."]
  }'

from openai import OpenAI

client = OpenAI(
    api_key="op-...",              # your OttoPort key
    base_url="https://ottoport.ai/api/v1",            # the only line that changes
)

resp = client.embeddings.create(
    model="bge-m3",
    input=["How do refunds work?", "Refunds are issued within 5 days."],
)
vectors = [item.embedding for item in resp.data]

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "op-...",               // your OttoPort key
  baseURL: "https://ottoport.ai/api/v1",             // the only line that changes
});

const resp = await client.embeddings.create({
  model: "bge-m3",
  input: ["How do refunds work?", "Refunds are issued within 5 days."],
});
const vectors = resp.data.map((item) => item.embedding);

Before you start

1. Create a key at /api-keys.
2. Top up credits — the smallest pack is $25.
3. Paste the request. It returns on the same request.

Full reference at /docs, or drive it from an agent with the MCP server and CLI.

Reference

Request parameters

The body POST https://ottoport.ai/api/v1/embeddings accepts for BGE-M3.

Field	Type	Required	Notes
model	string	Yes	Always bge-m3 for BGE-M3.
input	string \| array	Yes	One text, or an array of up to 256, each non-empty; 600,000 characters in all. One vector comes back per text, in order. Input tokens are the billed unit.
dimensions	integer	No	Fixed at 1024; any other value is refused.
encoding_format	string	No	`float` (default) or `base64`, little-endian float32.

Comparison

BGE-M3 direct vs through OttoPort

BGE-M3 is the same model either way — what changes is how many accounts, keys, and integrations you carry to reach it.

	BAAI direct	Via OttoPort
Accounts and keys	A BAAI account, plus one more for every other model you add	One OttoPort account and one key for the whole catalog
Request shape	BAAI's own SDK and response format	OpenAI-compatible — point baseURL at OttoPort and change the model string
Switching models	A new integration per provider	Replace bge-m3 with any other model ID in the catalog
Billing	A separate invoice and contract per provider	Prepaid credits, no subscription, one usage ledger
Provider outages	Your call fails	Configured backup routes are tried in order on provider failure

Embed queries and documents with the same model.

Batch up to 256 texts in one request.

Pick `dimensions` once and keep it.

FAQ

BGE-M3 API questions

Cost, keys, and model IDs for BGE-M3 on OttoPort.

How much does the BGE-M3 API cost?

On OttoPort, BGE-M3 costs $0.011 per 1M input tokens as of 2026-10-14. For scale, 10,000 passages: $0.055. Billing is prepaid usage with no subscription and no minimum, and the current rate is always published at ottoport.ai/pricing.

How do I get a BGE-M3 API key?

You do not need a BAAI key. Create an OttoPort account, generate a key at ottoport.ai/api-keys, and send it as "Authorization: Bearer <key>" to POST https://ottoport.ai/api/v1/embeddings with "model": "bge-m3". The same key reaches every other model in the catalog.

What is the model ID for BGE-M3?

The OttoPort model ID for BGE-M3 is bge-m3. Pass it as the model field on POST https://ottoport.ai/api/v1/embeddings, or list it programmatically from GET https://ottoport.ai/api/v1/models.

Can I call BGE-M3 with the OpenAI SDK?

Yes. Set the OpenAI SDK's base URL to https://ottoport.ai/api/v1, use an OttoPort API key, and pass bge-m3 as the model. Embeddings follow the OpenAI embeddings shape, so existing code works after changing two lines.

How many dimensions does BGE-M3 return?

1,024, always: BGE-M3 is not trained to be shortened, so a dimensions value other than 1024 is refused. Embed every text you compare with the same model.

Do I need a BAAI account to use BGE-M3?

No. OttoPort connects to BAAI on your behalf, so one OttoPort account covers BGE-M3 and every other model in the catalog — no per-provider contracts, keys, or invoices.

Alternatives

Other text embedding models on the same API

Same endpoint, same key — swapping model is a one-string change from bge-m3.

Voyage 4 API

Voyage AI

$0.066 per 1M input tokens

Voyage 4 Lite API

Voyage AI

$0.022 per 1M input tokens

Qwen3 Embedding 8B API

Qwen

$0.011 per 1M input tokens

Text Embedding 3 Small API

OpenAI

$0.022 per 1M input tokens

Text Embedding 3 Large API

OpenAI

$0.144 per 1M input tokens

Browse the full catalog — every model with its provider and per-unit price.