| license: apache-2.0 | |
| language: | |
| - en | |
| tags: | |
| - text-generation | |
| - mlx | |
| - quantized | |
| - zen | |
| - zenlm | |
| library_name: mlx | |
| base_model: zenlm/zen4-mini | |
| # Zen4 Mini (MLX) | |
| MLX quantization of Zen4 Mini for efficient inference on Apple Silicon. | |
| ## Model Details | |
| | Property | Value | | |
| |----------|-------| | |
| | **Parameters** | Zen4 Mini | | |
| | **Format** | MLX 4-bit quantized | | |
| | **Architecture** | Zen4 | | |
| | **License** | Apache 2.0 | | |
| | **Authors** | Zen LM Authors | | |
| ## Usage | |
| ```python | |
| from mlx_lm import load, generate | |
| model, tokenizer = load("zenlm/zen4-mini-MLX") | |
| response = generate(model, tokenizer, prompt="Hello, how can I help you?", max_tokens=512) | |
| print(response) | |
| ``` | |
| ## About | |
| Zen4 Mini is a compact, efficient language model from the Zen4 family, optimized for Apple Silicon via MLX quantization. | |
| **Developed by**: Zen LM Authors | |