quantized-models

1 article

All articles tagged with "quantized-models"

Running Transformers.js in Chrome: A Performance Review cover
-9 min read
Running Transformers.js in Chrome: A Performance Review
Transformers.js brings HuggingFace models into the browser via WebAssembly and WebGPU. After six months of running all-MiniLM-L6-v2 inside a Chrome extension for semantic search, here's what actually happens to performance, memory, and battery life.
ai-model-optimizationbrowser-based-aiquantized-modelswebassembly-wasmclient-side-ai
View all articles