EdgeQ runs Reka Edge on the Hexagon NPU of a Samsung S25, with reported time to first tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition → of 0.73s for a single image versus 1.6s for Gemma 4 E4B. Teams building on-device vision features get vendor-reported accuracy and latency figures.
Terms in this piece · Glossary
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
