FitMyLLM
BAAI/Dense

BAAIBGE-M3

BAAI's multilingual embedding model supporting 100+ languages.

embeddingDistilled
0.568B
Parameters
8K
Context length
0
Benchmarks
6
Quantizations
3.2M
HF downloads
Architecture
Dense
Released
2024-02-05
Layers
24
KV Heads
16
Head Dim
64
Family
embedding

Quantization Options

Context length:
QuantBitsVRAM @ 8KQuality
Q4_K_M4.89
1.2 GB
0.8 + 0.4 KV
good
Q5_K_S5.57
1.3 GB
0.9 + 0.4 KV
good
Q5_K_M5.7
1.3 GB
0.9 + 0.4 KV
good
Q6_K6.56
1.3 GB
1.0 + 0.4 KV
excellent
Q8_08.5
1.5 GB
1.1 + 0.4 KV
lossless
FP1616
2.0 GB
1.6 + 0.4 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
Ready to run this?Rent by the hour

Rent a GPU and run BGE-M3 now

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Run this model

Easiest way to get started·Beginners
Docs ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run bge-m3:latest

Ollama's default build of this model. For Q4_K_M specifically, pick a tag from ollama.com/library/bge-m3/tags.

Setup guide
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

Find the best GPU for BGE-M3

Build Hardware for BGE-M3

BAAI's multilingual embedding model supporting 100+ languages.

Spec sheet

BGE-M3 — 0.568B Dense.

Specifications
Parameters
0.568B
Architecture
Dense Transformer
Context Length
8K tokens
Capabilities
embedding
Release Date
2024-02-05
Provider
BAAI
Family
embedding
VRAM requirements
QuantBPWVRAMQuality
Q4_K_M4.890.8 GB94%
Q5_K_S5.570.9 GB96%
Q5_K_M5.70.9 GB96%
Q6_K6.561.0 GB97%
Q8_08.51.1 GB100%
FP16161.6 GB100%
Run command

Run BGE-M3 locally with Ollama — needs 0.8 GB VRAM at Q4_K_M:

$ollama run bge-m3:latest
Compatible GPUs
30 @ Q4_K_M