Skip to content

Latest commit

 

History

History
34 lines (22 loc) · 1.49 KB

changelog.md

File metadata and controls

34 lines (22 loc) · 1.49 KB

Change log

Release v2.0.0

  • Based on ggerganov/llama.cpp b3485 (https://github.com/ggerganov/llama.cpp/releases/tag/b3485)
  • Upgraded upstream tag enables Llama 3.1 in ollama
  • Native support for AmpereOne platform
  • Breaking change: due to changed weight type IDs it is now required to re-quantize models to Q8R16 and Q4_K_4 formats with current llama-quantize tool.

Release v1.2.7

Release v1.2.6

Release v1.2.5

Release v1.2.4

Release v1.2.3