Meltemi 7B Instruct (Chios dialect) — MLX 4-bit (fused)
This is an MLX-converted 4-bit Meltemi model with a Chios-dialect LoRA adapter fused in, so you can run it without specifying --adapter-path.
The training corpus was built from the pdf24_ocrPdf/ OCR PDFs and cleaned to contain only Greek / polytonic Greek letters (no Latin characters).
Usage
python -m mlx_lm chat --model YOUR_LOCAL_FOLDER_OR_HF_REPO --system-prompt "Μίλα πάντα στη χιώτικη διάλεκτο."
- Downloads last month
- 12
Model size
1B params
Tensor type
BF16
·
U32 ·
Hardware compatibility
Log In to add your hardware
4-bit
Model tree for ASiatras/Chios_Chatbot
Base model
ilsp/Meltemi-7B-Instruct-v1
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("ASiatras/Chios_Chatbot") prompt = "Write a story about Einstein" messages = [{"role": "user", "content": prompt}] prompt = tokenizer.apply_chat_template( messages, add_generation_prompt=True ) text = generate(model, tokenizer, prompt=prompt, verbose=True)