ms marco MiniLM L 12 v2 needs about 0.7 GB of memory at Q4_K (a 0.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 12 quants, from 0.0 GB (IQ4_XS) to 0.1 GB.
Cross‑encoder for MS Marco with 33 M parameters
ms marco MiniLM L 12 v2 is a BERT‑based cross‑encoder trained on the MS Marco Passage Ranking task. It encodes query‑passage pairs for ranking. The model is available in GGUF format for use with llama.cpp or CrispEmbed.
Good at
- Information retrieval ranking
- Embedding generation
- Cross‑encoder scoring
Limits
- No vision support
- Limited to 512‑token context
- Apache‑2.0 license
Best for
- Query‑passage ranking
- Re‑ranking search results
- Embedding similarity search
Who it’s for. Developers building search or recommendation systems that need efficient cross‑encoder ranking. The model is lightweight and can run on modest hardware.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ4_XS | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| IQ4_XS | 0.0 GB | 0.5 GB | Fits | Get file → | |
| IQ4_XS | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q4_K | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q8_0 | 0.0 GB | 0.5 GB | Fits | Get file → | |
| Q8_0 | 0.0 GB | 0.5 GB | Fits | Get file → | |
| — | 0.1 GB | 0.6 GB | Fits | Get file → | |
| — | 0.1 GB | 0.6 GB | Fits | Get file → |
For 8 GB VRAM, use IQ4_XS or Q4_K (file size 0.03 GB, memory ~0.7 GB). For 4 GB VRAM, IQ4_XS is recommended. If more precision is needed and 8 GB VRAM is available, use Q8_0 (file size 0.04 GB).
What GPU do you need for ms marco MiniLM L 12 v2?
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 (0.1 GB file, about 0.8 GB in use).
Which ms marco MiniLM L 12 v2 quant should you choose?
Q4_K (0.0 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 0.0 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 ms marco MiniLM L 12 v2
How much VRAM does ms marco MiniLM L 12 v2 need?
About 0.7 GB at Q4_K with an 8K context window. That is the 0.0 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 ms marco MiniLM L 12 v2 on an 8 GB GPU?
Yes, with the smaller files: the 0.1 GB file needs about 0.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of ms marco MiniLM L 12 v2 GGUF?
It depends on the quant: IQ4_XS is 0.0 GB, Q4_K is 0.0 GB, Q4_K is 0.0 GB, IQ4_XS is 0.0 GB, IQ4_XS is 0.0 GB, Q4_K is 0.0 GB and 6 more.
What context length does ms marco MiniLM L 12 v2 support?
The GGUF header lists a trained context length of 512 tokens (512). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.0 GB at 512.
Where can I download ms marco MiniLM L 12 v2 GGUF?
The files are on Hugging Face in the cstr/ms-marco-MiniLM-L-12-v2-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 ms marco MiniLM L 12 v2 use?
Hugging Face lists the license as apache-2.0. 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 this model on 4 GB VRAM?
Yes, the IQ4_XS quant uses about 0.7 GB of VRAM at 8k context, which fits within a 4 GB budget.
Does this model support image inputs?
No, it is a text‑only BERT cross‑encoder and does not process vision data.
What license applies to the GGUF files?
The GGUF files are distributed under the Apache‑2.0 license, identical to the upstream model.