Devstral-Small-2507-MLX-bf16 is a 24 billion parameter model from mistralai, optimized for Apple Silicon using MLX quantization. This bfloat16 version of Devstral-Small-2507 offers a 32768 token context length, making it suitable for tasks requiring substantial context processing on Apple hardware. Its primary differentiator is its specific optimization for MLX, enabling efficient local inference on Apple devices.
No reviews yet. Be the first to review!