Baai: BGE M3

Multilingual embedding model combining dense, sparse, and multi-vector retrieval

Model lab: Baai
Model family: bge

Model resources

Hugging Face

Modalitiestext → text

Context window8,192

Maximum output—

InputUSD 0.008 / 1M tokens

Try now

Quickstart

Use your Moduck API Key, set the API base URL to https://api.moduck.ai/v1, and set the model parameter to baai/bge-m3.

import os
import requests

base_url = "https://api.moduck.ai/v1"
api_key = os.environ["MODUCK_API_KEY"]
model_id = "baai/bge-m3"

response = requests.post(
    f"{base_url}/chat/completions",
    headers={
        "Authorization": f"Bearer {api_key}",
        "Content-Type": "application/json",
    },
    json={
        "model": model_id,
        "messages": [{"role": "user", "content": "Hello!"}],
        "stream": False,
    },
    timeout=60,
)
response.raise_for_status()
print(response.json())

Use the familiar OpenAI-style chat request format to make model integration easier.

BGE M3 API Pricing

Type	Price information
Input	USD 0.008 / 1M tokens

Performance

Measured on the Moduck gateway across live traffic, hour by hour over the last 24 hours. Time to first token is how long before anything appears; throughput is how fast the answer is written after that; end-to-end latency is the whole round trip. Figures are medians unless the card says otherwise.

Performance trends

Request throughput13.9 requests/minCurrent

Queue wait0.9 sCurrent

Completion time5.6 sCurrent

Request throughput

Current13.9 requests/min

Window average14.8 requests/min

Queue wait

Current0.9 s

Window average0.9 s

Completion time

Current5.6 s

Window average6.0 s

Tail completion time (P99)

Current13.2 s

Window average14.0 s

Retry rate

Current0.9%

Window average1.0%

Error rate

Current0.0%

Window average0.1%

Availability

Uptime is the share of time at least one provider serving this model responded. Availability is stricter: the share of requests actually served. A provider that answers but fails part of the traffic lowers availability without lowering uptime. Both cover the last three days.

Uptime100%At least one provider responded

Availability100%Requests actually served

Error rate0%Share of failed requests

Served requests, hour by hour

24 hours agoNow

BGE M3 Compare models

Model	Input price	Output price	Context window	Capabilities
BGE M3This page	USD 0.008 / 1M tokens	Not billed on output	8,192	Specialized model
Qwen3 Embedding 8B	USD 0.056 / 1M tokens	Not billed on output	32,768	Specialized model
Llama Nemotron Embed 1B v2	Price currently unavailable	Price currently unavailable	8,192	Specialized model

Model specifications

Input: text
Output: text
Context window: 8,192
Maximum output: —
Release date: 2024-01-30
Attachments: No
Reasoning: No
Tool calling: No
Temperature setting: No
Open weights: Yes

Frequently asked questions

What is BGE M3 for?

Multilingual embedding model combining dense, sparse, and multi-vector retrieval

What inputs and outputs does BGE M3 support?

Recorded inputs: text; outputs: text.

How is BGE M3 priced?

The pricing table shows the current recorded Moduck price. For example: Input USD 0.008 / 1M tokens. See the pricing table for all units, variants, and conditions.

What specifications are recorded for BGE M3?

Current model records list Context window 8,192 tokens.

What model parameter does the Quickstart use?

The example uses baai/bge-m3. This value comes from the page identifier and is not confirmed as a callable Moduck public ID.

Can I call this model through the Moduck API now?

This page has no verified Moduck API call for this model. A Quickstart example or reserved ID does not establish access; check the models available to your account.