Instructions to use Savyasaachin/gemma3n-lora-luna with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Savyasaachin/gemma3n-lora-luna with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-3n-e2b-it-unsloth-bnb-4bit")
model = PeftModel.from_pretrained(base_model, "Savyasaachin/gemma3n-lora-luna")

Transformers

How to use Savyasaachin/gemma3n-lora-luna with Transformers:

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

pipe = pipeline("text-generation", model="Savyasaachin/gemma3n-lora-luna")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Savyasaachin/gemma3n-lora-luna", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Savyasaachin/gemma3n-lora-luna with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Savyasaachin/gemma3n-lora-luna"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Savyasaachin/gemma3n-lora-luna",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Savyasaachin/gemma3n-lora-luna

SGLang

How to use Savyasaachin/gemma3n-lora-luna 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 "Savyasaachin/gemma3n-lora-luna" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Savyasaachin/gemma3n-lora-luna",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Savyasaachin/gemma3n-lora-luna" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Savyasaachin/gemma3n-lora-luna",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use Savyasaachin/gemma3n-lora-luna 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 Savyasaachin/gemma3n-lora-luna 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 Savyasaachin/gemma3n-lora-luna to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Savyasaachin/gemma3n-lora-luna to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Savyasaachin/gemma3n-lora-luna",
    max_seq_length=2048,
)

Docker Model Runner
How to use Savyasaachin/gemma3n-lora-luna with Docker Model Runner:
```
docker model run hf.co/Savyasaachin/gemma3n-lora-luna
```

gemma3n-lora-luna

This model is a fine-tuned version of unsloth/gemma-3n-e2b-it-unsloth-bnb-4bit on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.0176

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 1
eval_batch_size: 1
seed: 3407
gradient_accumulation_steps: 8
total_train_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 50
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1416437.4	0.0893	100	3.0176
68800.2375	0.1785	200	3.0176
54955.675	0.2678	300	3.0176
140025.7625	0.3571	400	3.0176
206484.625	0.4463	500	3.0176
1362432.3	0.5356	600	3.0176
3724982.4	0.6249	700	3.0176
1319876.5	0.7141	800	3.0176
3529.084	0.8034	900	3.0176
1194336.8	0.8927	1000	3.0176
363691.15	0.9819	1100	3.0176

Framework versions

PEFT 0.18.1
Transformers 4.56.2
Pytorch 2.9.0+cu126
Datasets 4.3.0
Tokenizers 0.22.2

Downloads last month: 1