bge m3 needs about 1.8 GB of memory at Q4_K (a 0.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 5 quants, from 0.4 GB (IQ4_XS) to 2.3 GB.
BGE-M3: Multi‑function, multi‑lingual embedding model
BGE-M3 is a 0.567B parameter model based on xlmr. It supports dense, multi‑vector, and sparse retrieval and handles up to 8194 context tokens. The model is licensed under MIT and is available in several GGUF quantisations.
Good at
- Dense and sparse retrieval
- Multi‑lingual support over 100 languages
- Hybrid retrieval pipelines
Limits
- No vision support
- Requires 8k context for full capability
- License is MIT
Best for
- Embedding generation for RAG
- Hybrid retrieval with BM25
- Multi‑vector retrieval in large corpora
Who it’s for. Engineers building retrieval‑augmented systems who need a single model for dense, sparse, and multi‑vector embeddings. The model is suitable for developers working with multilingual datasets and hybrid retrieval pipelines.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ4_XS | 0.4 GB | 1.8 GB | Fits | Get file → | |
| Q4_K | 0.5 GB | 1.8 GB | Fits | Get file → | |
| Q4_K | 0.5 GB | 1.8 GB | Fits | Get file → | |
| Q8_0 | 0.6 GB | 1.9 GB | Fits | Get file → | |
| — | 2.3 GB | 3.6 GB | Fits | Get file → |
For 8 GB VRAM use IQ4_XS or Q4_K, each ~0.45 GB file and ~1.8 GB memory at 8k context. For 16 GB VRAM Q8_0 (~0.61 GB file, ~1.9 GB memory) is suitable. The unknown quant (~2.27 GB file, ~3.6 GB memory) is for larger memory budgets.
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 the main file (2.3 GB file, about 3.6 GB in use).
Which bge m3 quant should you choose?
Q4_K (0.5 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
If memory is tight, the smallest file is IQ4_XS at 0.4 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about bge m3
How much VRAM does bge m3 need?
About 1.8 GB at Q4_K with an 8K context window. That is the 0.5 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 the smaller files: 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: IQ4_XS is 0.4 GB, Q4_K is 0.5 GB, Q4_K is 0.5 GB, Q8_0 is 0.6 GB, file is 2.3 GB.
What context length does bge m3 support?
The GGUF header lists a trained context length of 8,194 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 cstr/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, use IQ4_XS or Q4_K quantisations; each requires about 1.8 GB memory at 8k context.
Does BGE-M3 support image inputs?
No, the model only handles text embeddings.