BGE M3

Baai

baai/bge-m3Open Weights

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

Benchmark Pricing$0.01 / 1M in$0.00 / 1M out

Context: 8,192 tokens•Max Output: 4,096 tokens•Released: 2024-01-30

TECHNICAL SPECIFICATIONS

Context Window8,1928.2K tokens

Max Output4,0964.1K tokens

LicenseMITSelf-hostable

Input Modalities

Text

Function / Tool CallingNot Supported

Structured OutputsPrompt Based

MODEL CONTEXT PROTOCOL (MCP) READINESS

⚠ BGE M3 does not natively advertise tool-calling capabilities. When connecting to MCP servers, function calling must be handled via structured prompt injection.

MCP Context Window Recommendations

With a context capacity of 8,192 tokens, this model can handle compact tool sets. Consider keeping MCP server tool manifests minimal to preserve headroom.

API CALL INVOCATION EXAMPLE

// Example: Calling BGE M3 via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.BAAI_API_KEY,
});

const response = await client.chat.completions.create({
  model: "baai/bge-m3",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

console.log(response.choices[0].message.content);