https://www.solosoft.dev/es/tags/deep-learning/Recent content in Deep Learning on SoloSoftHugoes-esFri, 01 May 2026 00:00:00 +0000https://www.solosoft.dev/es/post/mlx-apple-silicon-2026/Fri, 01 May 2026 00:00:00 +0000https://www.solosoft.dev/es/post/mlx-apple-silicon-2026/<p>For years, machine learning on Macs meant one of two things: running PyTorch or TensorFlow through Apple&rsquo;s Metal Performance Shaders backend, or accepting that NVIDIA-optimized frameworks would never fully leverage Apple Silicon&rsquo;s capabilities. Both approaches left performance on the table. The unified memory architecture that makes M-series chips revolutionary for creative work went largely unused for ML.</p> <p>MLX changes this entirely. It is Apple&rsquo;s open-source ML framework, purpose-built for Apple Silicon. From the ground up, every optimization — lazy computation, unified memory access, neural engine integration — is designed for M-series hardware. The result is a framework that runs common ML workloads 2-3x faster on the same hardware compared to PyTorch through Metal, while using a cleaner, NumPy-inspired API.</p>https://www.solosoft.dev/es/post/mnn-mobile-inference-2026/Fri, 01 May 2026 00:00:00 +0000https://www.solosoft.dev/es/post/mnn-mobile-inference-2026/<p>Ejecutar modelos de deep learning en dispositivos moviles y de borde presenta desafios unicos: potencia de computo limitada, memoria restringida, sensibilidad de bateria y diversas arquitecturas de hardware. <strong>MNN</strong> (Mobile Neural Network) es la respuesta de Alibaba a estos desafios, un motor de inferencia ligero que lleva IA al borde con minima sobrecarga y maximo rendimiento.</p> <p>MNN impulsa mas de 30 aplicaciones de Alibaba, incluyendo Taobao (e-commerce), Youku (streaming de video) y varias herramientas empresariales. Ha sido probado en batalla a escala de miles de millones de usuarios, manejando desde vision artificial en tiempo real hasta LLMs en el dispositivo.</p>