BG

bge m3

BGEGGUF by CompendiumLabsUpdated 2025-07-17
Download on Hugging Face
Parameters567M
Context8K
Layers24
Architecturebert
Smallest quant1.2 GB
Quick answer

bge m3 needs about 2.5 GB of memory at F16 (a 1.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.2 GB (F16) to 2.3 GB (F32).

BGE M3 – 0.567 B parameter BERT model for embeddings

BGE M3 is a 0.567 B parameter BERT architecture designed for embedding generation. It supports a context length of 8192 tokens and is released under the MIT license. The model is available in F16 and F32 quantizations.

Good at

  • Embedding generation
  • Contextual similarity search
  • Semantic clustering

Limits

  • No vision or multimodal support
  • Requires 2.5 GB VRAM for F16 at 8k context
  • No active parameters or MoE

Best for

  • Search engine indexing
  • Document similarity tasks
  • Semantic search in small deployments

Who it’s for. Developers building embedding pipelines or semantic search solutions who need a lightweight BERT model with a large context window. The model is suitable for environments with limited GPU memory and a need for fast inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
F161.2 GB 2.5 GB
Fits Get file →
F322.3 GB 3.6 GB
Fits Get file →

For 8 GB VRAM, use the F16 quantization (1.16 GB file, ~2.5 GB VRAM at 8k context). If more memory is available, F32 (2.27 GB file, ~3.6 GB VRAM) offers higher precision.

What GPU do you need for bge m3?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is F32 (2.3 GB file, about 3.6 GB in use).

Which bge m3 quant should you choose?

F16 (1.2 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturebert
Parameters567M
Layers24
Trained context8,192 tokens
KV cache (fp16)96 KB per token

Source

Quantized byCompendiumLabs
FamilyBGE
Licensemit
Downloads (HF)3,500
Last updated2025-07-17

Frequently asked questions about bge m3

How much VRAM does bge m3 need?

About 2.5 GB at F16 with an 8K context window. That is the 1.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run bge m3 on an 8 GB GPU?

Yes, with F32: the 2.3 GB file needs about 3.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of bge m3 GGUF?

It depends on the quant: F16 is 1.2 GB, F32 is 2.3 GB.

What context length does bge m3 support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.8 GB at 8K and 0.8 GB at 8K.

Where can I download bge m3 GGUF?

The files are on Hugging Face in the CompendiumLabs/bge-m3-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does bge m3 use?

Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run BGE M3 on 8 GB VRAM?

Yes, the F16 quantization requires about 2.5 GB VRAM at 8k context, well within an 8 GB GPU.

Does BGE M3 support image embeddings?

No, BGE M3 is a text-only BERT model and does not support vision inputs.

Similar models

Browse all →

More from CompendiumLabs

All by CompendiumLabs →

More BGE models

All BGE →

Other tiny models

All tiny models →