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

Libraries

How to use second-state/embeddinggemma-300m-GGUF with sentence-transformers:

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("second-state/embeddinggemma-300m-GGUF")

sentences = [
    "That is a happy person",
    "That is a happy dog",
    "That is a very happy person",
    "Today is a sunny day"
]
embeddings = model.encode(sentences)

similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4]

llama-cpp-python

How to use second-state/embeddinggemma-300m-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="second-state/embeddinggemma-300m-GGUF",
	filename="embeddinggemma-300m-Q2_K.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use second-state/embeddinggemma-300m-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf second-state/embeddinggemma-300m-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf second-state/embeddinggemma-300m-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf second-state/embeddinggemma-300m-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf second-state/embeddinggemma-300m-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 second-state/embeddinggemma-300m-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf second-state/embeddinggemma-300m-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 second-state/embeddinggemma-300m-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf second-state/embeddinggemma-300m-GGUF:Q4_K_M

Use Docker

docker model run hf.co/second-state/embeddinggemma-300m-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use second-state/embeddinggemma-300m-GGUF with Ollama:
```
ollama run hf.co/second-state/embeddinggemma-300m-GGUF:Q4_K_M
```

Unsloth Studio new

How to use second-state/embeddinggemma-300m-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 second-state/embeddinggemma-300m-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 second-state/embeddinggemma-300m-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

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

Docker Model Runner
How to use second-state/embeddinggemma-300m-GGUF with Docker Model Runner:
```
docker model run hf.co/second-state/embeddinggemma-300m-GGUF:Q4_K_M
```

Lemonade

How to use second-state/embeddinggemma-300m-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull second-state/embeddinggemma-300m-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.embeddinggemma-300m-GGUF-Q4_K_M

List all available models

lemonade list

embeddinggemma-300m-Embedding-GGUF

Original Model

google/embeddinggemma-300m

Run with LlamaEdge

LlamaEdge version: v0.26.1 and above
Prompt template
- Prompt type: embedding
Context size: 2048
Embedding size: 128, 256, 512, 768

Run as LlamaEdge service

wasmedge --dir .:. --nn-preload default:GGML:AUTO:embeddinggemma-300m-f16.gguf \
  llama-api-server.wasm \
  --prompt-template embedding \
  --ctx-size 768 \
  --model-name embeddinggemma-300m

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
embeddinggemma-300m-Q2_K.gguf	Q2_K	2	212 MB	smallest, significant quality loss - not recommended for most purposes
embeddinggemma-300m-Q3_K_L.gguf	Q3_K_L	3	227 MB	small, substantial quality loss
embeddinggemma-300m-Q3_K_M.gguf	Q3_K_M	3	224 MB	very small, high quality loss
embeddinggemma-300m-Q3_K_S.gguf	Q3_K_S	3	218 MB	very small, high quality loss
embeddinggemma-300m-Q4_0.gguf	Q4_0	4	229 MB	legacy; small, very high quality loss - prefer using Q3_K_M
embeddinggemma-300m-Q4_K_M.gguf	Q4_K_M	4	236 MB	medium, balanced quality - recommended
embeddinggemma-300m-Q4_K_S.gguf	Q4_K_S	4	232 MB	small, greater quality loss
embeddinggemma-300m-Q5_0.gguf	Q5_0	5	242 MB	legacy; medium, balanced quality - prefer using Q4_K_M
embeddinggemma-300m-Q5_K_M.gguf	Q5_K_M	5	247 MB	large, very low quality loss - recommended
embeddinggemma-300m-Q5_K_S.gguf	Q5_K_S	5	243 MB	large, low quality loss - recommended
embeddinggemma-300m-Q6_K.gguf	Q6_K	6	260 MB	very large, extremely low quality loss
embeddinggemma-300m-Q8_0.gguf	Q8_0	8	329 MB	very large, extremely low quality loss - not recommended
embeddinggemma-300m-f16.gguf	f16	16	616 MB	very large, extremely low quality loss - not recommended

Quantized with llama.cpp b6397

Downloads last month: 606

GGUF

Model size

0.3B params

Architecture

gemma-embedding

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for second-state/embeddinggemma-300m-GGUF

Base model

google/embeddinggemma-300m

Quantized

(44)

this model