Instructions to use metricspace/embeddinggemma-300m-q4k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use metricspace/embeddinggemma-300m-q4k with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="metricspace/embeddinggemma-300m-q4k",
	filename="embeddinggemma-300M-Q4_K.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 metricspace/embeddinggemma-300m-q4k 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 metricspace/embeddinggemma-300m-q4k
# Run inference directly in the terminal:
llama cli -hf metricspace/embeddinggemma-300m-q4k

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf metricspace/embeddinggemma-300m-q4k
# Run inference directly in the terminal:
llama cli -hf metricspace/embeddinggemma-300m-q4k

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 metricspace/embeddinggemma-300m-q4k
# Run inference directly in the terminal:
./llama-cli -hf metricspace/embeddinggemma-300m-q4k

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 metricspace/embeddinggemma-300m-q4k
# Run inference directly in the terminal:
./build/bin/llama-cli -hf metricspace/embeddinggemma-300m-q4k

Use Docker

docker model run hf.co/metricspace/embeddinggemma-300m-q4k

LM Studio
Jan
Ollama
How to use metricspace/embeddinggemma-300m-q4k with Ollama:
```
ollama run hf.co/metricspace/embeddinggemma-300m-q4k
```

Unsloth Studio

How to use metricspace/embeddinggemma-300m-q4k 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 metricspace/embeddinggemma-300m-q4k 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 metricspace/embeddinggemma-300m-q4k to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for metricspace/embeddinggemma-300m-q4k to start chatting

Atomic Chat new
Docker Model Runner
How to use metricspace/embeddinggemma-300m-q4k with Docker Model Runner:
```
docker model run hf.co/metricspace/embeddinggemma-300m-q4k
```

Lemonade

How to use metricspace/embeddinggemma-300m-q4k with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull metricspace/embeddinggemma-300m-q4k

Run and chat with the model

lemonade run user.embeddinggemma-300m-q4k-{{QUANT_TAG}}

List all available models

lemonade list

EmbeddingGemma-300M — Q4_K GGUF (greppy runtime asset)

A Q4_K GGUF quantization of Google's EmbeddingGemma-300M, distributed as the embedded semantic-search backend of greppy, a local code-navigation tool for coding agents. greppy embeds this file directly in its binary and runs it on device (native Rust + vendored Metal/CUDA kernels — no llama.cpp runtime, Python, or model server) to power greppy semantic-search: finding code by meaning when the symbol name is unknown.

Base model: google/embeddinggemma-300m (includes the dense post-processing modules for sentence-transformers-compatible embeddings).
Quantization: Q4_K, from a pinned, byte-identical snapshot of cduk/embeddinggemma-300m-GGUF-with-dense-modules.
Use in greppy: query- and source-span embeddings for on-device retrieval.

License — Gemma Terms of Use (NOT Apache)

This model is a Model Derivative of Gemma and is provided under and subject to the Gemma Terms of Use, not an open-source license. By using or redistributing these files you agree, as an enforceable condition, that you will not use them for any purpose prohibited by the Gemma Prohibited Use Policy, and that any onward distribution carries these same restrictions, a copy of the Gemma Terms, and the notice below.

The binding terms, the Google notice, and the record of modifications are included in this repository:

GEMMA-TERMS.html
GEMMA-PROHIBITED-USE-POLICY.html
EMBEDDINGGEMMA-NOTICE.txt
EMBEDDINGGEMMA-MODIFICATIONS.txt
EMBEDDINGGEMMA-PROVENANCE.json

"Gemma" is a trademark of Google LLC. This is a community quantization for greppy and is not endorsed by or affiliated with Google.

Downloads last month: 88

GGUF

Model size

0.3B params

Architecture

gemma-embedding

Hardware compatibility

Inference Providers NEW

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

Model tree for metricspace/embeddinggemma-300m-q4k

Base model

google/embeddinggemma-300m

Quantized

(296)

this model