BG

bge m3 Q8 0

BGEGGUF by ggml-orgUpdated 2025-10-12
Download on Hugging Face
Parameters567M
Context8K
Layers24
Architecturebert
Smallest quant0.6 GB
Quick answer

bge m3 Q8 0 needs about 1.9 GB of memory at Q8_0 (a 0.6 GB file) with an 8K context window, so it fits on a 4 GB GPU.

BGE‑M3 Q8_0: 0.567B parameter embedding model with 8192‑token context

BGE‑M3 Q8_0 is a 0.567 billion‑parameter BERT‑based embedding model. It supports dense, multi‑vector, and sparse retrieval and can process up to 8192 tokens. The model is released under the MIT license.

Good at

  • Dense retrieval
  • Multi‑vector retrieval
  • Sparse retrieval

Limits

  • No vision support
  • Requires 1.9 GB VRAM at 8k context
  • MIT license only for non‑commercial use

Best for

  • Embedding generation for RAG pipelines
  • Hybrid retrieval with BM25
  • Re‑ranking after dense retrieval

Who it’s for. Developers building retrieval‑augmented systems who need a single model that handles dense, multi‑vector, and sparse retrieval. The model is suitable for applications requiring up to 8192‑token context and multi‑lingual support.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_00.6 GB 1.9 GB
Fits Get file →

The Q8_0 quantization file is 0.63 GB and needs about 1.9 GB VRAM at 8k context. For 8 GB VRAM, this quant is suitable. If VRAM is limited, consider a lower‑precision quant from the repository, if available.

What GPU do you need for bge m3 Q8 0?

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 Q8_0 (0.6 GB file, about 1.9 GB in use).

Which bge m3 Q8 0 quant should you choose?

Q8_0 (0.6 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

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 byggml-org
FamilyBGE
Base modelBAAI/bge-m3
Licensemit
Downloads (HF)22,770
Last updated2025-10-12

Frequently asked questions about bge m3 Q8 0

How much VRAM does bge m3 Q8 0 need?

About 1.9 GB at Q8_0 with an 8K context window. That is the 0.6 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 Q8 0 on an 8 GB GPU?

Yes, with Q8_0: the 0.6 GB file needs about 1.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of bge m3 Q8 0 GGUF?

It depends on the quant: Q8_0 is 0.6 GB.

What context length does bge m3 Q8 0 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 Q8 0 GGUF?

The files are on Hugging Face in the ggml-org/bge-m3-Q8_0-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 Q8 0 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 Q8_0 on 8 GB VRAM?

Yes, the Q8_0 quant requires approximately 1.9 GB VRAM at 8k context, which fits within an 8 GB GPU.

Does BGE‑M3 support multi‑lingual retrieval?

Yes, it supports more than 100 languages for embedding and sparse retrieval.

Is BGE‑M3 suitable for hybrid retrieval pipelines?

Yes, it can be used with BM25 or other methods for hybrid retrieval and re‑ranking.

Other GGUFs of bge-m3

Similar models

Browse all →

More from ggml-org

All by ggml-org →

More BGE models

All BGE →

Other tiny models

All tiny models →