Instructions to use vishalkhot/gemma-4b-distilled with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use vishalkhot/gemma-4b-distilled with Transformers:

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

pipe = pipeline("text-generation", model="vishalkhot/gemma-4b-distilled")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("vishalkhot/gemma-4b-distilled")
model = AutoModelForImageTextToText.from_pretrained("vishalkhot/gemma-4b-distilled")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use vishalkhot/gemma-4b-distilled with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "vishalkhot/gemma-4b-distilled"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "vishalkhot/gemma-4b-distilled",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/vishalkhot/gemma-4b-distilled

SGLang

How to use vishalkhot/gemma-4b-distilled 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 "vishalkhot/gemma-4b-distilled" \
    --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": "vishalkhot/gemma-4b-distilled",
		"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 "vishalkhot/gemma-4b-distilled" \
        --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": "vishalkhot/gemma-4b-distilled",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use vishalkhot/gemma-4b-distilled with Docker Model Runner:
```
docker model run hf.co/vishalkhot/gemma-4b-distilled
```

🧪 Gemma 4B — Distilled from Gemma 2

This repository contains a 4B-parameter distilled Gemma model, trained using knowledge distillation from a larger Gemma-2-9B-Instruct teacher.
The objective was to create a smaller, faster model that preserves strong instruction-following behavior while being practical for edge deployment and low-latency inference.

🔥 Overview

Teacher Model: google/gemma-2-9b-it
Student Model: vishalkhot/gemma-4b-distilled
Goal: Achieve near-teacher quality in a ~4B parameter footprint
Method: Logits-based distillation with temperature scaling, plus supervised tuning on curated instruction data

This model is ideal for:

Chat-based assistants
Reasoning over short/medium contexts
On-device inference (4B fits easily on a single modern GPU)
Serverless or low-cost API deployments

🧬 Distillation Process

Distillation was performed over a mixed instruction dataset, combining reasoning, multi-turn dialogue, tool-usage instructions, and diverse language tasks.
Training used a combination of:

🔹 KL divergence on teacher/student logits
🔹 Cross-entropy loss on reference outputs
🔹 Temperature scaling (T = 2–4)

Training Command Example

python distill.py \
  --teacher-model google/gemma-3-12b-it \
  --student-model google/gemma-3-4b-it\
  --train-file data/train.jsonl \
  --val-file data/val.jsonl \
  --output-dir gemma_4b_distilled \
  --batch-size 8 \
  --gradient-accumulation-steps 2 \
  --num-epochs 1 \
  --learning-rate 1e-5 \
  --warmup-steps 300 \
  --max-length 2048 \
  --temperature 2.5 \
  --alpha 0.5 \
  --save-steps 1000 \
  --eval-steps 200 \
  --logging-steps 10 \
  --bf16

Downloads last month: 1

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for vishalkhot/gemma-4b-distilled

Base model

google/gemma-3-12b-pt

Finetuned

google/gemma-3-12b-it

Finetuned

(338)

this model

vishalkhot
/

gemma-4b-distilled

🧪 Gemma 4B — Distilled from Gemma 2

🔥 Overview

🧬 Distillation Process

Training Command Example

Model tree for vishalkhot/gemma-4b-distilled

Dataset used to train vishalkhot/gemma-4b-distilled