AIThe Registerabout 2 hours ago
LLM runs on $10 ESP32 microcontroller
5 min read
A developer called SlvDev ran TinyStories, a 28.9 million-parameter model, on an ESP32-S3 microcontroller at 9.88 tokens per second. Quantization and per-layer-embedding offload cut memory needs from 60 MB to about 2 MB. The model costs under $10.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
