Instructions to use Chan-Y/TurkishReasoner-Gemma3-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Chan-Y/TurkishReasoner-Gemma3-1B with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-3-1b-it")
model = PeftModel.from_pretrained(base_model, "Chan-Y/TurkishReasoner-Gemma3-1B")

Transformers

How to use Chan-Y/TurkishReasoner-Gemma3-1B with Transformers:

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

pipe = pipeline("text-generation", model="Chan-Y/TurkishReasoner-Gemma3-1B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Chan-Y/TurkishReasoner-Gemma3-1B", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Chan-Y/TurkishReasoner-Gemma3-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Chan-Y/TurkishReasoner-Gemma3-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Chan-Y/TurkishReasoner-Gemma3-1B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Chan-Y/TurkishReasoner-Gemma3-1B

SGLang

How to use Chan-Y/TurkishReasoner-Gemma3-1B 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 "Chan-Y/TurkishReasoner-Gemma3-1B" \
    --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": "Chan-Y/TurkishReasoner-Gemma3-1B",
		"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 "Chan-Y/TurkishReasoner-Gemma3-1B" \
        --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": "Chan-Y/TurkishReasoner-Gemma3-1B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use Chan-Y/TurkishReasoner-Gemma3-1B 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 Chan-Y/TurkishReasoner-Gemma3-1B 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 Chan-Y/TurkishReasoner-Gemma3-1B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Chan-Y/TurkishReasoner-Gemma3-1B to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Chan-Y/TurkishReasoner-Gemma3-1B",
    max_seq_length=2048,
)

Docker Model Runner
How to use Chan-Y/TurkishReasoner-Gemma3-1B with Docker Model Runner:
```
docker model run hf.co/Chan-Y/TurkishReasoner-Gemma3-1B
```

TurkishReasoner-Gemma3-1B

Model Description

TurkishReasoner-Gemma1B is a lightweight Turkish reasoning model fine-tuned from Google's Gemma3-1B. Despite its compact size, this model delivers impressive reasoning capabilities in Turkish, making it ideal for deployment in resource-constrained environments while maintaining high-quality step-by-step reasoning.

Key Features

Built on Google's efficient Gemma3-1B foundation
Fine-tuned specifically for Turkish reasoning tasks
Optimized for deployment on devices with limited resources
Delivers structured reasoning with clearly formatted solutions
Efficient text-only processing for reasoning tasks
32K token context window

Technical Specifications

Base Model: Google/Gemma3-1B
Parameters: 1 billion
Input: Text only
Hardware Requirements: ~4GB VRAM
Training Infrastructure: NVIDIA T4 GPU

Usage

This model is ideal for applications requiring reasoning capabilities in resource-constrained environments:

Mobile applications with Turkish reasoning capabilities
Educational tools for deployment on standard consumer hardware
Embedded systems requiring compact reasoning abilities
Local inference on personal computers with limited GPU resources

Example Usage

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

base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-3-1b-it")
model = PeftModel.from_pretrained(base_model, "Chan-Y/TurkishReasoner-Gemma3-1B").to("cuda")
tokenizer = AutoTokenizer.from_pretrained("unsloth/gemma-3-1b-it")

pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_new_tokens=512,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
)

messages = [
    {"role": "system", "content": """Sen kullanıcıların isteklerine Türkçe cevap veren bir asistansın ve sana bir problem verildi.
Problem hakkında düşün ve çalışmanı göster.
Çalışmanı <start_working_out> ve <end_working_out> arasına yerleştir.
Sonra, çözümünü <SOLUTION> ve </SOLUTION> arasına yerleştir.
Lütfen SADECE Türkçe kullan."""},
    {"role": "user", "content": "121'in karekökü kaçtır?"},
]

response = pipe(messages, return_full_text=False)[0]["generated_text"]
print(response)