Instructions to use keisuke-miyako/bge-m3-doc-r4-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use keisuke-miyako/bge-m3-doc-r4-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="keisuke-miyako/bge-m3-doc-r4-gguf",
	filename="bge-m3-doc-r4-q8_0.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 keisuke-miyako/bge-m3-doc-r4-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 keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0
# Run inference directly in the terminal:
llama cli -hf keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0
# Run inference directly in the terminal:
llama cli -hf keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0

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 keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0

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 keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0

Use Docker

docker model run hf.co/keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0

LM Studio
Jan
Ollama
How to use keisuke-miyako/bge-m3-doc-r4-gguf with Ollama:
```
ollama run hf.co/keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0
```

Unsloth Studio

How to use keisuke-miyako/bge-m3-doc-r4-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 keisuke-miyako/bge-m3-doc-r4-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 keisuke-miyako/bge-m3-doc-r4-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for keisuke-miyako/bge-m3-doc-r4-gguf to start chatting

Atomic Chat new
Docker Model Runner
How to use keisuke-miyako/bge-m3-doc-r4-gguf with Docker Model Runner:
```
docker model run hf.co/keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0
```

Lemonade

How to use keisuke-miyako/bge-m3-doc-r4-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull keisuke-miyako/bge-m3-doc-r4-gguf:Q8_0

Run and chat with the model

lemonade run user.bge-m3-doc-r4-gguf-Q8_0

List all available models

lemonade list

bge-m3-doc-r4 (GGUF q8_0)

Fine-tuned on 4D doc document embeddings. Dataset: keisuke-miyako/doc-2026-0615

Usage (llama.cpp)

./llama-embedding -m bge-m3-doc-r4-q8_0.gguf --prompt "your query"

Benchmarks

Relevance	Min	Max	Average
`3`	`0.31`	`0.84`	`0.64`
`2`	`0.04`	`0.82`	`0.60`
`1`	`0.06`	`0.77`	`0.53`
`0`	`-0.11`	`0.61`	`0.19`

avg. spread: 0.45
lv. 3 vs 2 is separated by 0.04
lv. 3 vs 1 is separated by 0.11
lv. 0 is well separated at 0.19 (-0.01) 👍🏻

Threshold	Positive	Negative	Gap
0.54	`0.97`	`0.31`	`0.65`
0.55	`0.95`	`0.29`	`0.65`
0.56	`0.93`	`0.26`	`0.67`
0.57	`0.90`	`0.23`	`0.67`
0.58	`0.88`	`0.18`	`0.69`
0.59	`0.83`	`0.15`	`0.67`
0.60	`0.78`	`0.12`	`0.65`
0.61	`0.74`	`0.09`	`0.64`

gap peak: 0.58

Downloads last month: 45

GGUF

Model size

0.6B params

Architecture

bert

Hardware compatibility

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for keisuke-miyako/bge-m3-doc-r4-gguf

Base model

BAAI/bge-m3

Quantized

(276)

this model