Instructions to use gururaser/qwen3-4b-bilimkurgu with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use gururaser/qwen3-4b-bilimkurgu with Transformers:

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

pipe = pipeline("text-generation", model="gururaser/qwen3-4b-bilimkurgu")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("gururaser/qwen3-4b-bilimkurgu")
model = AutoModelForCausalLM.from_pretrained("gururaser/qwen3-4b-bilimkurgu", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use gururaser/qwen3-4b-bilimkurgu with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gururaser/qwen3-4b-bilimkurgu"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "gururaser/qwen3-4b-bilimkurgu",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/gururaser/qwen3-4b-bilimkurgu

SGLang

How to use gururaser/qwen3-4b-bilimkurgu 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 "gururaser/qwen3-4b-bilimkurgu" \
    --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": "gururaser/qwen3-4b-bilimkurgu",
		"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 "gururaser/qwen3-4b-bilimkurgu" \
        --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": "gururaser/qwen3-4b-bilimkurgu",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use gururaser/qwen3-4b-bilimkurgu with Docker Model Runner:
```
docker model run hf.co/gururaser/qwen3-4b-bilimkurgu
```

Qwen3 4B Bilimkurgu - Türkçe MMLU Benchmark & Model Kartı

Geliştirici: gururaser
Lisans: cc-by-nc-4.0
Eğitildiği Taban Model: unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit
Eğitim Veri Seti: gururaser/ithaki-bilimkurgu-klasikleri

📈 Türkçe MMLU Benchmark Test Sonuçları

Model, alibayram/yapay_zeka_turkce_mmlu_model_cevaplari test veri kümesi kullanılarak Türkçe MMLU benchmark değerlendirmesine tabi tutulmuştur.

📊 Türkçe MMLU Benchmark Karşılaştırma Tablosu

Model Türü	Model Adı	Genel Başarı (%)	Doğru / Toplam Soru	Test Süresi (sn)
Base Model	`unsloth/qwen3-4b-instruct-2507-unsloth-bnb-4bit`	%26.21	1625 / 6200	218.4s
Fine-Tuned Model	`gururaser/qwen3-4b-bilimkurgu`	%34.35	2130 / 6200	251.23s

🧪 Test Metodolojisi ve Detaylar

Değerlendirme Scripti: alibayram/yapay_zeka_turkce_mmlu_bolum_sonuclari/olcum.py mantığı temel alınmıştır.
Anlamsal Karşılaştırma: paraphrase-multilingual-mpnet-base-v2 modeli ile şık eşleştirme ve opsiyonel harf eşleme.
Ortam: Google Colab A100 GPU (4-bit NF4 Quantization)
Test Seti: alibayram/yapay_zeka_turkce_mmlu_model_cevaplari

Downloads last month: 8

Safetensors

Model size

4B params

Tensor type

BF16

gururaser
/

qwen3-4b-bilimkurgu

Qwen3 4B Bilimkurgu - Türkçe MMLU Benchmark & Model Kartı

📈 Türkçe MMLU Benchmark Test Sonuçları

📊 Türkçe MMLU Benchmark Karşılaştırma Tablosu

🧪 Test Metodolojisi ve Detaylar

Datasets used to train gururaser/qwen3-4b-bilimkurgu