Google DeepMindLaunchAI ModelsNovember 16, 2026

Google updates small Gemma 4 tokenizers to fix nonsense output

Read original source →huggingface.co

Google pushed tokenizer fixes to Gemma 4 E4B, E2B and 12B, addressing an issue that produced nonsense output when the models hit certain tokens. Whether the fix reaches existing GGUF quantizations is unclear.

1 source

More stories today

Open the live feed