Explains why a 975B-parameter MoE multimodal model can't run locally in .NET via ONNX Runtime GenAI due to size, architecture, and conversion limits.
A technical guide on using the GeoDeep Python package to run AI object detection models on Maxar's satellite imagery, focusing on earthquake-affected areas.
Mozilla's experiment enabling AI model inference directly in Firefox Web extensions using Transformers.js and ONNX, with a practical example.
Learn to optimize Sentence Transformers models for faster inference using Hugging Face Optimum, ONNX Runtime, and dynamic quantization.
Learn to accelerate Vision Transformer (ViT) models using quantization with Hugging Face Optimum and ONNX Runtime for improved latency.
Learn to optimize Hugging Face Transformers models for GPU inference using Optimum and ONNX Runtime to reduce latency.
Learn to optimize Hugging Face Transformers models using Optimum and ONNX Runtime for faster inference with dynamic quantization.
Learn how to use Hugging Face Optimum and ONNX Runtime to apply static quantization to a DistilBERT model, achieving ~3x latency improvements.
