§ feed · storyline
Google AI releases faster Gemma 4 inference with multi-token
Google AI releases multi-token prediction drafters for Gemma 4, enabling up to 3x faster inference via speculative decoding with no reported quality loss.
Google AI released Multi-Token Prediction (MTP) drafters for Gemma 4, offering up to 3x faster inference without quality loss using speculative decoding.
§ sources1 publication · timeline below