Instructions to use smarttasks/mxbai-embed-large-v1-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use smarttasks/mxbai-embed-large-v1-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="smarttasks/mxbai-embed-large-v1-GGUF",
	filename="mxbai-embed-large-v1-Q4_K_M.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use smarttasks/mxbai-embed-large-v1-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use smarttasks/mxbai-embed-large-v1-GGUF with Ollama:
```
ollama run hf.co/smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M
```

Unsloth Studio

How to use smarttasks/mxbai-embed-large-v1-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for smarttasks/mxbai-embed-large-v1-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for smarttasks/mxbai-embed-large-v1-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for smarttasks/mxbai-embed-large-v1-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use smarttasks/mxbai-embed-large-v1-GGUF with Docker Model Runner:
```
docker model run hf.co/smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M
```

Lemonade

How to use smarttasks/mxbai-embed-large-v1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull smarttasks/mxbai-embed-large-v1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.mxbai-embed-large-v1-GGUF-Q4_K_M

List all available models

lemonade list

Usage note: retrieval query prefix

For retrieval tasks, prepend this instruction to queries (documents get no prefix):

Represent this sentence for searching relevant passages:

The MTEB score below (SciFact 0.7388 nDCG@10) was measured with this query prefix. For similarity/clustering tasks, no prefix is needed. Context window is 512 tokens.

mxbai-embed-large-v1 — Embedding GGUF (quantization-verified)

Quantized embedding model in GGUF, served in --embedding mode via llama.cpp. This is an encoder — it outputs vectors, not text. It is validated for retrieval quality and quantization fidelity, not chat behavior.

Files

mxbai-embed-large-v1-Q4_K_M.gguf (215.8 MB)
mxbai-embed-large-v1-Q5_K_M.gguf (245.7 MB)
mxbai-embed-large-v1-Q8_0.gguf (358.1 MB)

Quantization drift (vs f16)

Mean cosine similarity of embeddings vs the f16 baseline. 1.0 = identical.

Quant	Mean cosine	Min cosine	Verdict
Q4_K_M	0.99304	0.99006	excellent (>0.99)
Q5_K_M	0.99746	0.99578	excellent (>0.99)
Q8_0	0.99982	0.99965	excellent (>0.99)

Per-domain fidelity at Q4_K_M (which content types the quant preserves best):

Domain	Mean cosine	Min
finance	0.99234	0.99006
code	0.99263	0.99148
medical	0.9929	0.99151
long_form	0.99307	0.99263
legal	0.99313	0.99077
science	0.99315	0.99214
everyday	0.99333	0.99114
short_queries	0.99377	0.99318

Retrieval sanity (lightweight)

Built-in 12-query retrieval check (no external corpus): top-1 accuracy 1.0, MRR 1.0. healthy (top-1 >= 0.9)

Retrieval (MTEB)

Standardized MTEB retrieval scores (main metric, usually nDCG@10 — higher is better). These are comparable across models on the MTEB leaderboard.

Task	Score
SciFact	0.7388

Metric: main_score (retrieval tasks: nDCG@10). Measured on the Q8_0 quant served via llama.cpp.

Dense-retrieval mode. These scores are for standard single-vector dense retrieval (what llama.cpp serves). Models like BGE-M3 that also support sparse/multi-vector (ColBERT) modes score higher in hybrid setups — that capability isn't exercised here, so compare this number against other models' dense scores, not hybrid ones.

What this is NOT

This card carries no safety, red-team, or viewpoint scores: those do not apply to an embedding model. For chat-model governance cards, see the SmartTasks text-LLM line.

Downloads last month: -

GGUF

Model size

0.3B params

Architecture

bert

Hardware compatibility

4-bit

5-bit

8-bit

Model tree for smarttasks/mxbai-embed-large-v1-GGUF

Base model

mixedbread-ai/mxbai-embed-large-v1

Quantized

(15)

this model