Instructions to use Tuguberk/Qwen3.5-2B-Turkish-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with Transformers:

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

pipe = pipeline("text-generation", model="Tuguberk/Qwen3.5-2B-Turkish-SFT")
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("Tuguberk/Qwen3.5-2B-Turkish-SFT")
model = AutoModelForImageTextToText.from_pretrained("Tuguberk/Qwen3.5-2B-Turkish-SFT")
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]:]))

llama-cpp-python

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Tuguberk/Qwen3.5-2B-Turkish-SFT",
	filename="Qwen3.5-2B.BF16-mmproj.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Use Docker

docker model run hf.co/Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

LM Studio
Jan

vLLM

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Tuguberk/Qwen3.5-2B-Turkish-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Tuguberk/Qwen3.5-2B-Turkish-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

SGLang

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT 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 "Tuguberk/Qwen3.5-2B-Turkish-SFT" \
    --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": "Tuguberk/Qwen3.5-2B-Turkish-SFT",
		"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 "Tuguberk/Qwen3.5-2B-Turkish-SFT" \
        --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": "Tuguberk/Qwen3.5-2B-Turkish-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with Ollama:
```
ollama run hf.co/Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M
```

Unsloth Studio

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT 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 Tuguberk/Qwen3.5-2B-Turkish-SFT 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 Tuguberk/Qwen3.5-2B-Turkish-SFT to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Tuguberk/Qwen3.5-2B-Turkish-SFT to start chatting

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with Docker Model Runner:
```
docker model run hf.co/Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M
```

Lemonade

How to use Tuguberk/Qwen3.5-2B-Turkish-SFT with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.5-2B-Turkish-SFT-Q4_K_M

List all available models

lemonade list

Qwen3.5-2B-Turkish-SFT

Qwen3.5-2B base modeli üzerine Türkçe instruction-following verisiyle fine-tune edilmiş bir dil modelidir.

Model Detayları


Base model	unsloth/Qwen3.5-2B
Fine-tuning yöntemi	LoRA (bf16)
LoRA rank	16
Dataset	AlicanKiraz0/Turkish-SFT-Dataset-v1.0
Dataset boyutu	5.579 örnek
Epoch	3
Learning rate	2e-4
Training loss	1.47 → 0.82
Framework	Unsloth + TRL
GPU	NVIDIA L4 (22GB)

Kullanım

Transformers

from transformers import AutoModelForCausalLM, AutoProcessor
import torch

model_name = "Tuguberk/Qwen3.5-2B-Turkish-SFT"

processor = AutoProcessor.from_pretrained(model_name)
tokenizer = processor.tokenizer

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

messages = [
    {"role": "system", "content": "Sen yardımcı bir Türkçe asistansın."},
    {"role": "user",   "content": "Türkiye hakkında kısa bir bilgi verir misin?"},
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=False,
)
inputs = tokenizer(text, return_tensors="pt").to(model.device)

outputs = model.generate(
    inputs["input_ids"],
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.9,
    do_sample=True,
)

response = tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True)
print(response)

Ollama (GGUF)

ollama run hf.co/Tuguberk/Qwen3.5-2B-Turkish-SFT:Q4_K_M

llama.cpp

# Metin modeli
llama-cli -hf Tuguberk/Qwen3.5-2B-Turkish-SFT --jinja

# Multimodal
llama-mtmd-cli -hf Tuguberk/Qwen3.5-2B-Turkish-SFT --jinja

Mevcut Formatlar

Dosya	Boyut	Açıklama
`Qwen3.5-2B.BF16-mmproj.gguf`	~4.5GB	En yüksek kalite
`Qwen3.5-2B.Q8_0.gguf`	~2.7GB	Yüksek kalite
`Qwen3.5-2B.Q4_K_M.gguf`	~1.5GB	Önerilen — kalite/boyut dengesi
`Qwen3.5-2B.Q2_K.gguf`	~0.8GB	En küçük, düşük kalite

Tavsiye: Çoğu kullanım için Q4_K_M yeterlidir.

Dataset

AlicanKiraz0/Turkish-SFT-Dataset-v1.0 — genel amaçlı Türkçe instruction-following verisi.

Format: system / user / assistant üçlüsü, ChatML şablonuyla işlenmiştir.

Evaluation

Benchmark değerlendirmeleri lm-evaluation-harness-turkish kullanılarak yapılmıştır.

Benchmark	Base Model	Fine-tuned	Fark
Turkish MMLU (`mmlu_tr_v0.2`)	47.29%	47.20%	−0.09
GSM8K-TR strict (`gsm8k_tr-v0.2`)	38.04%	41.38%	+3.34
TruthfulQA (`truthfulqa_v0.2`)	47.45%	49.08%	+1.63
ARC-TR normalized (`arc_tr-v0.2`)	37.03%	38.23%	+1.20
HellaSwag-TR normalized (`hellaswag_tr-v0.2`)	38.74%	40.27%	+1.53
Winogrande-TR (`winogrande_tr-v0.2`)	49.53%	49.61%	+0.08

Önemli bulgular:

Matematik muhakemesinde (GSM8K-TR) +3.34 puanlık belirgin iyileşme
Doğruluk ve akıl yürütme gerektiren benchmark'larda (TruthfulQA, ARC, HellaSwag) tutarlı iyileşme
Genel bilgi (MMLU) korunmuş — catastrophic forgetting gözlemlenmedi
Yalnızca 5.579 örnekle elde edilmiş sonuçlar

Sınırlamalar

5.5K örnekle eğitilmiştir — büyük ölçekli veriyle eğitilmiş modellerle kıyaslanamaz
Derin matematik ve zincirleme muhakeme görevlerinde sınırlı performans
Yalnızca Türkçe instruction-following için optimize edilmiştir

LoRA Adaptörleri

Sadece LoRA ağırlıklarını kullanmak için: Tuguberk/Qwen3.5-2B-Turkish-SFT-LoRA

Lisans

Bu model, base model ile aynı lisansı taşımaktadır: Apache 2.0

Downloads last month: 476

Safetensors

Model size

2B params

Tensor type

F32

BF16

Model tree for Tuguberk/Qwen3.5-2B-Turkish-SFT

Base model

Qwen/Qwen3.5-2B-Base

Finetuned

Qwen/Qwen3.5-2B

Finetuned

unsloth/Qwen3.5-2B

Quantized

(10)

this model

Tuguberk
/

Qwen3.5-2B-Turkish-SFT

Qwen3.5-2B-Turkish-SFT

Model Detayları

Kullanım

Transformers

Ollama (GGUF)

llama.cpp

Mevcut Formatlar

Dataset

Evaluation

Sınırlamalar

LoRA Adaptörleri

Lisans

Model tree for Tuguberk/Qwen3.5-2B-Turkish-SFT

Dataset used to train Tuguberk/Qwen3.5-2B-Turkish-SFT

Space using Tuguberk/Qwen3.5-2B-Turkish-SFT 1