Instructions to use alibidaran/Gemma2_Farsi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alibidaran/Gemma2_Farsi with Transformers:

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

pipe = pipeline("text-generation", model="alibidaran/Gemma2_Farsi")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("alibidaran/Gemma2_Farsi")
model = AutoModelForCausalLM.from_pretrained("alibidaran/Gemma2_Farsi")

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use alibidaran/Gemma2_Farsi with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alibidaran/Gemma2_Farsi"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "alibidaran/Gemma2_Farsi",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/alibidaran/Gemma2_Farsi

SGLang

How to use alibidaran/Gemma2_Farsi 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 "alibidaran/Gemma2_Farsi" \
    --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": "alibidaran/Gemma2_Farsi",
		"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 "alibidaran/Gemma2_Farsi" \
        --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": "alibidaran/Gemma2_Farsi",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use alibidaran/Gemma2_Farsi with Docker Model Runner:
```
docker model run hf.co/alibidaran/Gemma2_Farsi
```

fine tuning

by arshiaafshani - opened Dec 28, 2024

Discussion

arshiaafshani

Dec 28, 2024

•

edited Dec 28, 2024

hi
how can I fine tune this model? and also is there any way to change it's algorithms?

alibidaran

Owner Dec 28, 2024

Hi,
You need to use the TRL library to fine-tune it because it requires several memory optimization methods to use them on Colab and T4 GPUs.
For the second question, I don't understand what you mean about changing the algorithm.

arshiaafshani

Dec 28, 2024

•

edited Dec 28, 2024

I meant changing the process of generating a prompt, the codes that contain the generation a result. Can I have your Telegram username?

alibidaran

Owner Dec 29, 2024

Yes you can do that. You change prompt style or you can use your own way

arshiaafshani

Dec 29, 2024

Thanks !

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