Instructions to use nazimali/Mistral-Nemo-Kurdish with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nazimali/Mistral-Nemo-Kurdish with Transformers:

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

pipe = pipeline("text-generation", model="nazimali/Mistral-Nemo-Kurdish")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nazimali/Mistral-Nemo-Kurdish")
model = AutoModelForCausalLM.from_pretrained("nazimali/Mistral-Nemo-Kurdish")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nazimali/Mistral-Nemo-Kurdish with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/nazimali/Mistral-Nemo-Kurdish

SGLang

How to use nazimali/Mistral-Nemo-Kurdish 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 "nazimali/Mistral-Nemo-Kurdish" \
    --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": "nazimali/Mistral-Nemo-Kurdish",
		"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 "nazimali/Mistral-Nemo-Kurdish" \
        --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": "nazimali/Mistral-Nemo-Kurdish",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use nazimali/Mistral-Nemo-Kurdish 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 nazimali/Mistral-Nemo-Kurdish 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 nazimali/Mistral-Nemo-Kurdish to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for nazimali/Mistral-Nemo-Kurdish to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="nazimali/Mistral-Nemo-Kurdish",
    max_seq_length=2048,
)

Docker Model Runner
How to use nazimali/Mistral-Nemo-Kurdish with Docker Model Runner:
```
docker model run hf.co/nazimali/Mistral-Nemo-Kurdish
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Continued pre-training on mistralai/Mistral-Nemo-Instruct-2407 using the Kurdish wiki dataset with unsloth. This model should be further fine-tuned since the pre-training was to improve Kurdish language understanding. It's a quantized model using bitsandbytes so that it uses less memory. See bitsandbytes documentation.

There isn't a standard or even a good Kurdish metric to evaluate the model (that I could find). Will make it my next project to create an evaluation so that there's a reproducible baseline for Kurdish.

Will look into a multi-GPU training setup so don't have to wait all day for results. Would like to train it with both Kurmanji and Sorani.

Use

Should be fine-tuned further for a specific task. See instruction fine-tuned model nazimali/Mistral-Nemo-Kurdish-Instruct.

Training

Transformers 4.44.2
1 NVIDIA A100 80GB PCIe
Duration 6h 31m 4s

{
  "total_flos": 4121524790259794000,
  "train/epoch": 1,
  "train/global_step": 1960,
  "train/grad_norm": 3.1958093643188477,
  "train/learning_rate": 0,
  "train/loss": 1.2108,
  "train_loss": 1.256846008738693,
  "train_runtime": 23227.1752,
  "train_samples_per_second": 2.7,
  "train_steps_per_second": 0.084
}

Pre-training data:

nazimali/kurdish-wikipedia-articles
- Dataset number of rows: 63,076
- Filtered columns title, text
  - Must have at least 1 character
Number of rows used for training: 62,720

Training prompt format:

training_prompt = """Gotara Wikipedia
### Sernav: {}

### Gotar:
{}"""

Downloads last month: 17

Safetensors

Model size

12B params

Tensor type

BF16

Model tree for nazimali/Mistral-Nemo-Kurdish

Base model

mistralai/Mistral-Nemo-Base-2407

Finetuned

mistralai/Mistral-Nemo-Instruct-2407

Finetuned

(209)

this model

Finetunes

1 model

Quantizations

6 models

nazimali
/

Mistral-Nemo-Kurdish

Use

Training

Pre-training data:

Training prompt format:

Model tree for nazimali/Mistral-Nemo-Kurdish

Dataset used to train nazimali/Mistral-Nemo-Kurdish

Space using nazimali/Mistral-Nemo-Kurdish 1