llama.cpp
Add quantization recipes from custom recipe files
#21070
Open
Overview
Commits
30
Changes
View On GitHub
Commits
  • Add unit test coverage for llama_tensor_get_type
    bartowski1182 committed 29 days ago
  • Fix merge conflicts, add more schemas
    bartowski1182 committed 29 days ago
  • clang formatter changes
    bartowski1182 committed 29 days ago
  • Trailing whitespace
    bartowski1182 committed 29 days ago
  • Update name
    bartowski1182 committed 29 days ago
  • Start rebase
    bartowski1182 committed 29 days ago
  • Updating files with upstream changes prior to rebase
    bartowski1182 committed 29 days ago
  • Changes needed from rebase
    bartowski1182 committed 29 days ago
  • Update attn_qkv schema, change throw behaviour
    bartowski1182 committed 29 days ago
  • Fix merge conflicts
    bartowski1182 committed 29 days ago
  • White space
    bartowski1182 committed 29 days ago
  • Update with latest changes to state counters
    bartowski1182 committed 29 days ago
  • Revert accidental personal CLAUDE.md changes
    bartowski1182 committed 29 days ago
  • Change quotation mark
    bartowski1182 committed 29 days ago
  • Reuse metadata.name since we have it
    bartowski1182 committed 29 days ago
  • Move test-only stuff out of llama-quant.cpp
    bartowski1182 committed 29 days ago
  • Hide the regex functionality back in llama-quant.cpp, use a unique pointer to a new struct 'compiled_tensor_type_patterns' which contains the patterns
    bartowski1182 committed 29 days ago
  • Merge branch 'ggml-org:master' into llama-quant-refactor
    bartowski1182 committed 28 days ago
  • cont : inital deslop guidelines
    ggerganov committed 28 days ago
  • Merge branch 'ggml-org:master' into llama-quant-refactor
    bartowski1182 committed 26 days ago
  • Cleanup based on review comments
    bartowski1182 committed 25 days ago
  • Continue cleanup
    bartowski1182 committed 25 days ago
  • Small cleanup
    bartowski1182 committed 25 days ago
  • Merge branch 'ggml-org:master' into llama-quant-refactor
    bartowski1182 committed 25 days ago
  • Merge branch 'ggml-org:master' into llama-quant-refactor
    bartowski1182 committed 18 days ago
  • Manually set proper ordering of tensors, mostly applies to gemma
    bartowski1182 committed 16 days ago
  • Add quantization recipes
    bartowski1182 committed 16 days ago
  • Fix compile warnings
    bartowski1182 committed 16 days ago
  • Update tests
    bartowski1182 committed 16 days ago
  • Add specific attention v counters and a way to group them like in legacy code, update category names to match existing
    bartowski1182 committed 15 days ago
FAQ Terms Privacy Refunds Impressum
Loading