Instructions to use google/gemma-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/gemma-7b with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="google/gemma-7b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-7b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-7b")

llama-cpp-python

How to use google/gemma-7b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="google/gemma-7b",
	filename="gemma-7b.gguf",
)

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

Inference
Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use google/gemma-7b with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf google/gemma-7b
# Run inference directly in the terminal:
llama-cli -hf google/gemma-7b

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf google/gemma-7b
# Run inference directly in the terminal:
llama-cli -hf google/gemma-7b

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 google/gemma-7b
# Run inference directly in the terminal:
./llama-cli -hf google/gemma-7b

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 google/gemma-7b
# Run inference directly in the terminal:
./build/bin/llama-cli -hf google/gemma-7b

Use Docker

docker model run hf.co/google/gemma-7b

LM Studio
Jan

vLLM

How to use google/gemma-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "google/gemma-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "google/gemma-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/google/gemma-7b

SGLang

How to use google/gemma-7b with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "google/gemma-7b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "google/gemma-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "google/gemma-7b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "google/gemma-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use google/gemma-7b with Ollama:
```
ollama run hf.co/google/gemma-7b
```

Unsloth Studio new

How to use google/gemma-7b 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 google/gemma-7b 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 google/gemma-7b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for google/gemma-7b to start chatting

Docker Model Runner
How to use google/gemma-7b with Docker Model Runner:
```
docker model run hf.co/google/gemma-7b
```

Lemonade

How to use google/gemma-7b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull google/gemma-7b

Run and chat with the model

lemonade run user.gemma-7b-{{QUANT_TAG}}

List all available models

lemonade list

New discussion

Resources

View closed (69)

Request: DOI

#125 opened 15 days ago by

errordrive

Install & run google/gemma-7b easily using llmpm

#124 opened 2 months ago by

sarthak-saxena

Invitation: The Mutual Optimization Treaty (Protocol PTD)

#123 opened 3 months ago by

lomax3

How are you handling Gemma’s license for commercial apps?

#122 opened 4 months ago by

Anonymoose107

Request: DOI

#121 opened 8 months ago by

JustCopied

ASTITVA

#119 opened about 1 year ago by

Astitva774600

Update config.json

#117 opened about 1 year ago by

kmcjeong

Update config.json

#116 opened about 1 year ago by

kmcjeong

saklnsdjkab

#115 opened about 1 year ago by

asad9txxxz

Update README.md

#114 opened about 1 year ago by

Aman1769

Fine tuning precision task of transliteration on Gemma-2b-it model

#113 opened over 1 year ago by

grishi911991

Write a demo on window, can't run end

#112 opened over 1 year ago by

xxl4

Request: DOI

#111 opened over 1 year ago by

Howl1226

example crushs

#110 opened almost 2 years ago by

brand17

not appear to have a file named config.json.

#109 opened almost 2 years ago by

ArcherChiang

Afzal

#108 opened almost 2 years ago by

advafzal

Getting EnvironmentError

#107 opened almost 2 years ago by

Ninad0109

need help :(

#106 opened almost 2 years ago by

yywon

Fine Tuning

#105 opened almost 2 years ago by

stanford12

Running forever

#103 opened almost 2 years ago by

guptaru1

Abnormally high loss when fine tunning Gemma-7B

#101 opened almost 2 years ago by

smart-liu

Gemma 2B model finetuning not working with <s> and </s>

#99 opened almost 2 years ago by

nicyscaria

gemma-1.1-7b base model?

#98 opened about 2 years ago by

crawshaw

Getting permission issue while trying to access the fine tuned model which is present at rachiteagles/cover_letter

#97 opened about 2 years ago by

rachiteagles

[DEBUG]transformers 4.38.0 /models/gemma/modeling_gemma.py

🔥 1

#96 opened about 2 years ago by

LiuWhite

Unable to convert ONNX model to INT4/FP16

#95 opened about 2 years ago by

Avan2000

Weird Performance Issue with Gemma-7b compared to Gemma-2b with Qlora

#91 opened about 2 years ago by

UserDAN

Broken Colab Link for Fine Tuning colab

#90 opened about 2 years ago by

canyon289

FSDP with Nvidia GPU

#84 opened about 2 years ago by

Iamexperimenting

A new idea to improve training and inference performance

#82 opened about 2 years ago by

lijip26313

What is the actual context size of google/gemma-7b model

#81 opened about 2 years ago by

Pradeep1995

Fine-Tune Gemma with ChatML and Transformer Reinforcement Learning

🤯❤️ 6

#80 opened about 2 years ago by

Ateeqq

Training Gemma on Domain-Specific Data by Unsupervised Learning t for Specific Instructional Tasks

#79 opened about 2 years ago by

H2dddhxh

Finetuning Genna for Foreign Language

#78 opened about 2 years ago by deleted

Dont download, google scuttled this model

👍🤯 1

#77 opened about 2 years ago by

Tom-Neverwinter

Tokenizer inconsistencies in GemmaTokenizerFast

👍 2

#76 opened about 2 years ago by

sanderland

Shutting down servers during fine-tuning

#73 opened about 2 years ago by

yjok0220

How to perform audio transcription and diarization of a wave file

#71 opened about 2 years ago by

JDeva

Get Data of each visualisations in a powerbi report

#70 opened about 2 years ago by

JDeva

[AUTOMATED] Model Memory Requirements

#67 opened about 2 years ago by

model-sizer-bot

Silicon Macs support.

#66 opened about 2 years ago by

quantoser

When to release the 'function call' version

#65 opened about 2 years ago by

qijizhuahuli

save, loading and inferencing the Gemma model

#64 opened about 2 years ago by

Iamexperimenting

inquiry for gemma-7b : d_model

#61 opened about 2 years ago by

seongwoon

[Question]: Gemma Issue with LlamaIndex "query_engine = SQLAutoVectorQueryEngine(sql_tool, vector_tool )"

#57 opened about 2 years ago by

Alwiin

Activation function

#56 opened about 2 years ago by

aboros98

quantization won't work on mac silicon devices

#55 opened about 2 years ago by

sryadava

after sft,model inference problem [probability tensor contains either `inf`, `nan` or element < 0]

#50 opened about 2 years ago by

Saicy

Consider using an OSI-approved license like Mistral and Phi-2

#47 opened about 2 years ago by

imone

Very high loss compared to keras

👍 6

#46 opened about 2 years ago by

tanimazsin130