alexgusevski/Eva-4B-mlx-fp16
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Jan 12, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm
alexgusevski/Eva-4B-mlx-fp16 is a 4 billion parameter language model converted to the MLX format from FutureMa/Eva-4B. This model is specifically designed for efficient deployment and inference on Apple Silicon, leveraging the MLX framework. It provides a readily available MLX-optimized version of the Eva-4B architecture for local machine learning applications.
Loading preview...
Overview
This model, alexgusevski/Eva-4B-mlx-fp16, is a 4 billion parameter language model that has been converted to the MLX format. The conversion was performed from the original FutureMa/Eva-4B model using mlx-lm version 0.29.1. This makes it suitable for efficient inference on Apple Silicon devices.
Key Capabilities
- MLX Optimization: Specifically formatted for use with the MLX framework, enabling optimized performance on Apple Silicon.
- Direct Use: Ready for immediate use with the
mlx-lmlibrary, requiring minimal setup. - Chat Template Support: Includes support for applying chat templates, facilitating conversational AI applications.
Good For
- Local Inference: Ideal for developers and researchers looking to run a 4B parameter model locally on Apple Silicon hardware.
- MLX Ecosystem Integration: Seamlessly integrates into projects utilizing the MLX machine learning framework.
- Experimentation: Provides a convenient way to experiment with the Eva-4B architecture in an MLX environment.