Skip to content
Live
ITHome · AI·· 1 hour ago

Google engineers fit the 270M parameter Gemma 3 model with 270 million parameters into a printer to run

Brief overview

Google DeepMind engineer Omar Sanseverino ran a 270M parameter Gemma 3 270M model with 14.1MB of memory, taking about 60 seconds per token, by compiling the Transformer inference engine and the quantized model into a PostScript print job and executing the computation using the printer's built-in interpreter.

Source: ITHome · AI · ithome.com