What hardware do I need to run BGE-M3?

At its most compressed (Int8) it needs roughly 0.7GB of VRAM, and about 1.2GB for good quality. VRAM figures are approximate and depend on context length and settings.

Is BGE-M3 free for commercial use?

Yes. BGE-M3 is licensed under MIT, which permits commercial use with no meaningful conditions.

Can I run BGE-M3 on Apple Silicon?

Yes. BGE-M3 has builds optimised for Apple Silicon, through MLX or GGUF on a Mac.

Does BGE-M3 run on CPU?

Yes, BGE-M3 can run on the CPU, though generation is slower than on a GPU.

What is BGE-M3's context window?

BGE-M3 has a context window of 8,192 tokens, about 8k.

BGE-M3

Name: BGE-M3
Author: BAAI

BAAI · Embedding · 568M · 8k context · Released 30 January 2024

Commercial use permitted Open weights Runs on CPU Apple Silicon

A multilingual embedding model that is unusual for doing dense, sparse, and multi-vector retrieval from a single model, across more than 100 languages, with an 8192-token input. MIT licensed and a common default for local RAG.

Strengths

Dense, sparse, and multi-vector retrieval in one model
Over 100 languages, useful for multilingual document sets
MIT licensed, and a long 8192-token input

Weaknesses

Heavier and slower than tiny embedders like nomic-embed
Multi-vector retrieval adds storage and retrieval complexity
A general model, so a domain-tuned embedder may beat it on narrow tasks

Hardware requirements

Quantisation	Approx. VRAM	Notes
FP16	~1.2GB	Full precision, runs comfortably on any GPU or CPU
Int8	~0.7GB	Lighter, minimal quality loss for retrieval

Also runs on CPU (slower). Optimised builds available for Apple Silicon.

What you'd need to run this

Roughly what a machine to run this would need, at up to three levels of quality. Memory is the deciding factor.

Minimum to run it

Int8 · ~0.7GB needed

One 12GB GPU

NVIDIA GeForce RTX 3060 12GB

or a Mac or mini-PC with unified memory, if you prefer no discrete GPU, Mac mini M5 Pro .

At least 32GB of system RAM alongside the card.

around £700–£1,100

What else 12GB runs →

Best quality

FP16 · ~1.2GB needed

One 12GB GPU

NVIDIA GeForce RTX 3060 12GB

or a Mac or mini-PC with unified memory, if you prefer no discrete GPU, Mac mini M5 Pro .

At least 32GB of system RAM alongside the card.

around £700–£1,100

What else 12GB runs →

Licence

MIT — read the licence

Benchmarks

Benchmark	Score	Source	As of
MTEB Multilingual Mean (Task)	59.56	Qwen3-Embedding-0.6B model card (comparison table, reported by Qwen)	June 2025

How it compares

How this model’s reported scores sit against other models we cover, on the same benchmarks. This model is highlighted.

MTEB Multilingual Mean (Task)

higher is better

Qwen3-Embedding 0.6B 64.33

Qwen3-Embedding-0.6B model card · June 2025
BGE-M3 59.56

Qwen3-Embedding-0.6B model card (comparison table, reported by Qwen) · June 2025

BGE-M3: common questions

What hardware do I need to run BGE-M3?: At its most compressed (Int8) it needs roughly 0.7GB of VRAM, and about 1.2GB for good quality. VRAM figures are approximate and depend on context length and settings.
Is BGE-M3 free for commercial use?: Yes. BGE-M3 is licensed under MIT, which permits commercial use with no meaningful conditions.
Can I run BGE-M3 on Apple Silicon?: Yes. BGE-M3 has builds optimised for Apple Silicon, through MLX or GGUF on a Mac.
Does BGE-M3 run on CPU?: Yes, BGE-M3 can run on the CPU, though generation is slower than on a GPU.
What is BGE-M3's context window?: BGE-M3 has a context window of 8,192 tokens, about 8k.

Availability

Official page
Hugging Face
ollama run bge-m3

Recommended for

Multilingual retrieval for RAG systems
Document sets that mix languages
A capable default when you are not sure which embedder to use

Related models

Run it with

Catalogue entry last verified 30 July 2026. Specifications change; verify anything you are about to spend money on.