llama.cpp
Gemma 4: move some computations to BF16
#21451
Closed
Overview
Commits
4
Changes
View On GitHub
Commits
  • YATF (Yet Another Tokenizer Fix) for Gemma 4. With tests!
    pwilkin committed 4 days ago
  • Remove unnecessary hash from update script.
    pwilkin committed 4 days ago
  • gemma4 bf16
    pwilkin committed 4 days ago
  • Add comparison script, incorporate @ggerganov's precision fix
    pwilkin committed 4 days ago
FAQ Terms Privacy Refunds Impressum
Loading