Instructions to use shanjivkr/vit-gemma-modelNew with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use shanjivkr/vit-gemma-modelNew with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-2b-it-bnb-4bit")
model = PeftModel.from_pretrained(base_model, "shanjivkr/vit-gemma-modelNew")

Transformers

How to use shanjivkr/vit-gemma-modelNew with Transformers:

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

pipe = pipeline("text-generation", model="shanjivkr/vit-gemma-modelNew")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("shanjivkr/vit-gemma-modelNew", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use shanjivkr/vit-gemma-modelNew with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/shanjivkr/vit-gemma-modelNew

SGLang

How to use shanjivkr/vit-gemma-modelNew 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 "shanjivkr/vit-gemma-modelNew" \
    --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": "shanjivkr/vit-gemma-modelNew",
		"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 "shanjivkr/vit-gemma-modelNew" \
        --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": "shanjivkr/vit-gemma-modelNew",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use shanjivkr/vit-gemma-modelNew 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 shanjivkr/vit-gemma-modelNew 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 shanjivkr/vit-gemma-modelNew to start chatting

Using HuggingFace Spaces for Unsloth

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

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="shanjivkr/vit-gemma-modelNew",
    max_seq_length=2048,
)

Docker Model Runner
How to use shanjivkr/vit-gemma-modelNew with Docker Model Runner:
```
docker model run hf.co/shanjivkr/vit-gemma-modelNew
```

vit-gemma-modelNew / handler.py

shanjivkr

Upload handler.py with huggingface_hub

c9398de verified 3 months ago

raw

history blame contribute delete

1.06 kB


	import torch
	from transformers import AutoTokenizer, AutoModelForCausalLM
	from peft import PeftModel

	class EndpointHandler():
	def __init__(self, path=""):
	# Load the base model and the LoRA adapters
	tokenizer = AutoTokenizer.from_pretrained(path)
	model = AutoModelForCausalLM.from_pretrained(
	path,
	device_map="auto",
	torch_dtype=torch.bfloat16
	)
	self.tokenizer = tokenizer
	self.model = model
	self.device = "cuda" if torch.cuda.is_available() else "cpu"

	def __call__(self, data):
	inputs = data.get("inputs", data)
	input_ids = self.tokenizer(inputs, return_tensors="pt").input_ids.to(self.device)

	with torch.no_grad():
	outputs = self.model.generate(
	input_ids,
	max_new_tokens=128,
	temperature=0.7,
	top_p=0.9
	)

	prediction = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
	return [{"generated_text": prediction}]