Instructions to use ngquocvinh/EmbeddingGemma-2-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ngquocvinh/EmbeddingGemma-2-GGUF with sentence-transformers:

from sentence_transformers import SentenceTransformer

model = SentenceTransformer("ngquocvinh/EmbeddingGemma-2-GGUF")

sentences = [
    "The weather is lovely today.",
    "It's so sunny outside!",
    "He drove to the stadium."
]
embeddings = model.encode(sentences)

similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use ngquocvinh/EmbeddingGemma-2-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 ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ngquocvinh/EmbeddingGemma-2-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 ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ngquocvinh/EmbeddingGemma-2-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 ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ngquocvinh/EmbeddingGemma-2-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 ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M

Use Docker

docker model run hf.co/ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use ngquocvinh/EmbeddingGemma-2-GGUF with Ollama:
```
ollama run hf.co/ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use ngquocvinh/EmbeddingGemma-2-GGUF with Docker Model Runner:
```
docker model run hf.co/ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M
```

Lemonade

How to use ngquocvinh/EmbeddingGemma-2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ngquocvinh/EmbeddingGemma-2-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.EmbeddingGemma-2-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

ngquocvinh commited on 4 days ago

Commit

5dcf67b

verified ·

1 Parent(s): da064ac

Add Q3_K_S after text embedding smoke pass

Browse files

Files changed (5) hide show

.gitattributes +1 -0
EmbeddingGemma-2-Q3_K_S.gguf +3 -0
SHA256SUMS.txt +1 -0
reproducibility/artifact-measurements.tsv +1 -0
reproducibility/manifest.md +8 -0

.gitattributes CHANGED Viewed

@@ -45,3 +45,4 @@ EmbeddingGemma-2-IQ1_M.gguf filter=lfs diff=lfs merge=lfs -text
 EmbeddingGemma-2-Q1_0.gguf filter=lfs diff=lfs merge=lfs -text
 EmbeddingGemma-2-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
 EmbeddingGemma-2-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text

 EmbeddingGemma-2-Q1_0.gguf filter=lfs diff=lfs merge=lfs -text
 EmbeddingGemma-2-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
 EmbeddingGemma-2-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
+EmbeddingGemma-2-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text

EmbeddingGemma-2-Q3_K_S.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:a6c72c2febc19a36afa8de82fb7e6cac8a45a2e7a20922612b2cc7c8b19270f6
+size 142545536

SHA256SUMS.txt CHANGED Viewed

@@ -3,6 +3,7 @@ c4e59a215a94e4ab077e1a10c43df85ab1c63addc8fdf5515a5ee9b99c8a15f4  EmbeddingGemma
 4953a59ae0d8a725052407f458101ef81d8acb6a95f3293f4949ff3f594ef18a  EmbeddingGemma-2-Q1_0.gguf
 67124cbba678cce4664bf9af72839a88ab12a9083821545c56847f41a54e96a1  EmbeddingGemma-2-Q2_K.gguf
 a5c902b15458c4886d6f1296bad71fd8084d7ac626cb8713778cf2af5e1b5a71  EmbeddingGemma-2-Q3_K_M.gguf
 ea25550b050fa8826707c5e44659a7c18259606dcc6ba449f771e3c75cf4e663  EmbeddingGemma-2-Q4_K_M.gguf
 4a9b92529b930be9f6cc858821b5d6ceea46c9631a05bd27f33a6942f46b63f6  EmbeddingGemma-2-Q4_K_S.gguf
 22f79226a02bfa2dd840414f342a004f93b3cb85e498923a5e089f03f6410b1f  EmbeddingGemma-2-Q5_K_M.gguf

 4953a59ae0d8a725052407f458101ef81d8acb6a95f3293f4949ff3f594ef18a  EmbeddingGemma-2-Q1_0.gguf
 67124cbba678cce4664bf9af72839a88ab12a9083821545c56847f41a54e96a1  EmbeddingGemma-2-Q2_K.gguf
 a5c902b15458c4886d6f1296bad71fd8084d7ac626cb8713778cf2af5e1b5a71  EmbeddingGemma-2-Q3_K_M.gguf
+a6c72c2febc19a36afa8de82fb7e6cac8a45a2e7a20922612b2cc7c8b19270f6  EmbeddingGemma-2-Q3_K_S.gguf
 ea25550b050fa8826707c5e44659a7c18259606dcc6ba449f771e3c75cf4e663  EmbeddingGemma-2-Q4_K_M.gguf
 4a9b92529b930be9f6cc858821b5d6ceea46c9631a05bd27f33a6942f46b63f6  EmbeddingGemma-2-Q4_K_S.gguf
 22f79226a02bfa2dd840414f342a004f93b3cb85e498923a5e089f03f6410b1f  EmbeddingGemma-2-Q5_K_M.gguf

reproducibility/artifact-measurements.tsv CHANGED Viewed

@@ -11,3 +11,4 @@ EmbeddingGemma-2-IQ1_M.gguf	103040640	0.103041	PASS_EMBEDDING_CPU
 EmbeddingGemma-2-Q1_0.gguf	66163328	0.066163	PASS_EMBEDDING_CPU
 EmbeddingGemma-2-Q5_K_S.gguf	210670208	0.210670	PASS_EMBEDDING_CPU
 EmbeddingGemma-2-Q4_K_S.gguf	178164352	0.178164	PASS_EMBEDDING_CPU

 EmbeddingGemma-2-Q1_0.gguf	66163328	0.066163	PASS_EMBEDDING_CPU
 EmbeddingGemma-2-Q5_K_S.gguf	210670208	0.210670	PASS_EMBEDDING_CPU
 EmbeddingGemma-2-Q4_K_S.gguf	178164352	0.178164	PASS_EMBEDDING_CPU
+EmbeddingGemma-2-Q3_K_S.gguf	142545536	0.142546	PASS_EMBEDDING_CPU

reproducibility/manifest.md CHANGED Viewed

@@ -120,3 +120,11 @@ The fixed task evaluation uses the full MTEB STSBenchmark `test` split and the u
 - Quantization command: `llama-quantize --imatrix calibration/embeddinggemma-2/embeddinggemma-2-sts.imatrix.gguf embeddinggemma-2-text-BF16.gguf EmbeddingGemma-2-Q4_K_S.gguf Q4_K_S 8`.
 - CPU `llama-server --embedding --pooling mean -ngl 0` smoke passed with finite normalized 768-dimensional text vectors.
 - Build cgroup: `MemoryMax=32G`, `MemorySwapMax=0`; peak memory `926449664` bytes and swap peak `0` bytes.

 - Quantization command: `llama-quantize --imatrix calibration/embeddinggemma-2/embeddinggemma-2-sts.imatrix.gguf embeddinggemma-2-text-BF16.gguf EmbeddingGemma-2-Q4_K_S.gguf Q4_K_S 8`.
 - CPU `llama-server --embedding --pooling mean -ngl 0` smoke passed with finite normalized 768-dimensional text vectors.
 - Build cgroup: `MemoryMax=32G`, `MemorySwapMax=0`; peak memory `926449664` bytes and swap peak `0` bytes.
+- Q4_K_S public Hub commit: `da064ac951ec91745594fe8699f3044842fc92ae`; remote LFS SHA256 and size match the local file.
+## Incremental experiment: Q3_K_S
+- Artifact: `EmbeddingGemma-2-Q3_K_S.gguf`; SHA256 `a6c72c2febc19a36afa8de82fb7e6cac8a45a2e7a20922612b2cc7c8b19270f6`; decimal size `0.142546 GB`.
+- Quantization command: `llama-quantize --imatrix calibration/embeddinggemma-2/embeddinggemma-2-sts.imatrix.gguf embeddinggemma-2-text-BF16.gguf EmbeddingGemma-2-Q3_K_S.gguf Q3_K_S 8`.
+- CPU `llama-server --embedding --pooling mean -ngl 0` smoke passed with finite normalized 768-dimensional text vectors.
+- Build cgroup: `MemoryMax=32G`, `MemorySwapMax=0`; peak memory `870584320` bytes and swap peak `0` bytes.