Instructions to use TURKCELL/Turkcell-LLM-7b-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TURKCELL/Turkcell-LLM-7b-v1 with Transformers:

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

pipe = pipeline("text-generation", model="TURKCELL/Turkcell-LLM-7b-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TURKCELL/Turkcell-LLM-7b-v1")
model = AutoModelForCausalLM.from_pretrained("TURKCELL/Turkcell-LLM-7b-v1")
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

vLLM

How to use TURKCELL/Turkcell-LLM-7b-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TURKCELL/Turkcell-LLM-7b-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TURKCELL/Turkcell-LLM-7b-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TURKCELL/Turkcell-LLM-7b-v1

SGLang

How to use TURKCELL/Turkcell-LLM-7b-v1 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 "TURKCELL/Turkcell-LLM-7b-v1" \
    --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": "TURKCELL/Turkcell-LLM-7b-v1",
		"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 "TURKCELL/Turkcell-LLM-7b-v1" \
        --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": "TURKCELL/Turkcell-LLM-7b-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TURKCELL/Turkcell-LLM-7b-v1 with Docker Model Runner:
```
docker model run hf.co/TURKCELL/Turkcell-LLM-7b-v1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Turkcell-LLM-7b-v1

This model is an extended version of a Mistral-based Large Language Model (LLM) for Turkish. It was trained on a cleaned Turkish raw dataset containing 5 billion tokens. The training process involved using the DORA method initially. Following this, we utilized Turkish instruction sets created from various open-source and internal resources for fine-tuning with the LORA method.

Model Details

Base Model: Mistral 7B based LLM
Tokenizer Extension: Specifically extended for Turkish
Training Dataset: Cleaned Turkish raw data with 5 billion tokens, custom Turkish instruction sets
Training Method: Initially with DORA, followed by fine-tuning with LORA

DORA Configuration

lora_alpha: 128
lora_dropout: 0.05
r: 64
target_modules: "all-linear"

LORA Fine-Tuning Configuration

lora_alpha: 128
lora_dropout: 0.05
r: 256
target_modules: "all-linear"

Usage Examples


from transformers import AutoModelForCausalLM, AutoTokenizer

device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained("TURKCELL/Turkcell-LLM-7b-v1")
tokenizer = AutoTokenizer.from_pretrained("TURKCELL/Turkcell-LLM-7b-v1")

messages = [
    {"role": "user", "content": "Türkiye'nin başkenti neresidir?"},
]

encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")

eos_token = tokenizer("<|im_end|>",add_special_tokens=False)["input_ids"][0]

model_inputs = encodeds.to(device)
model.to(device)

generated_ids = model.generate(model_inputs, 
                               max_new_tokens=1024, 
                               do_sample=True, 
                               eos_token_id=eos_token)
                               
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])