Instructions to use mradermacher/translategemma-4b-it-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mradermacher/translategemma-4b-it-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mradermacher/translategemma-4b-it-GGUF", dtype="auto")

llama-cpp-python

How to use mradermacher/translategemma-4b-it-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="mradermacher/translategemma-4b-it-GGUF",
	filename="translategemma-4b-it.IQ4_XS.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use mradermacher/translategemma-4b-it-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mradermacher/translategemma-4b-it-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mradermacher/translategemma-4b-it-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 mradermacher/translategemma-4b-it-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mradermacher/translategemma-4b-it-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 mradermacher/translategemma-4b-it-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mradermacher/translategemma-4b-it-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 mradermacher/translategemma-4b-it-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mradermacher/translategemma-4b-it-GGUF:Q4_K_M

Use Docker

docker model run hf.co/mradermacher/translategemma-4b-it-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use mradermacher/translategemma-4b-it-GGUF with Ollama:
```
ollama run hf.co/mradermacher/translategemma-4b-it-GGUF:Q4_K_M
```

Unsloth Studio new

How to use mradermacher/translategemma-4b-it-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 mradermacher/translategemma-4b-it-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 mradermacher/translategemma-4b-it-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mradermacher/translategemma-4b-it-GGUF to start chatting

Docker Model Runner
How to use mradermacher/translategemma-4b-it-GGUF with Docker Model Runner:
```
docker model run hf.co/mradermacher/translategemma-4b-it-GGUF:Q4_K_M
```

Lemonade

How to use mradermacher/translategemma-4b-it-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mradermacher/translategemma-4b-it-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.translategemma-4b-it-GGUF-Q4_K_M

List all available models

lemonade list

4B not good enough for image transacription + translation in a single query?

by metalglot - opened Apr 12

Discussion

metalglot

Apr 12

•

edited Apr 12

I've got this working for image processing, but with low quantization levels 4B returns nonsense, but at a high quantitization is just returns english.

Where as 12B/27B returns the transcribed text fully translated.

Has anyone else run into this issue?

Think the workaround for 4B not transcribing + translating in one query is to separate it into two queries? One to transcribe the contents of the image, then a second to translate the transcription?

metalglot changed discussion title from 4B not good enough for transacription + translation in a single query? to 4B not good enough for image transacription + translation in a single query? Apr 12

nicoboss

Apr 14

Try it. Transcribing plus translating seams a bit much all at once for a 4B model. Generally 4B might simply be too small for such complex tasks.

metalglot

Apr 14

I have tried it - it's just transcribing if it is a high enough quant, lower quant it returns nonsense. IMO it needs a transcription pass, then a separate translation query, rather than both in one query..

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment