Add quantization recipes from custom recipe files #21070
bartowski1182
wants to merge
30 commits into
ggml-org:master
from
bartowski1182:quant-recipes
Add unit test coverage for llama_tensor_get_type
99119cea
Fix merge conflicts, add more schemas
a3ff1940
clang formatter changes
363e6d3f
Trailing whitespace
0a89cdae
Update name
86103e7e
Start rebase
86273028
Updating files with upstream changes prior to rebase
6e414fc2
Changes needed from rebase
2015dea8
Update attn_qkv schema, change throw behaviour
544745c0
Fix merge conflicts
182cbe5d
White space
506a4900
Update with latest changes to state counters
3948227d
Revert accidental personal CLAUDE.md changes
aa8d567e
Change quotation mark
d2586d50
Reuse metadata.name since we have it
3fe55f10
Move test-only stuff out of llama-quant.cpp
8ebfe03f
Hide the regex functionality back in llama-quant.cpp, use a unique po…
4a2f648d
Merge branch 'ggml-org:master' into llama-quant-refactor
64d6c881
cont : inital deslop guidelines
d576ae32
Merge branch 'ggml-org:master' into llama-quant-refactor
3adf377a
Cleanup based on review comments
0b3cc323
Continue cleanup
b85a7c8c
Small cleanup
c7aa761b
Merge branch 'ggml-org:master' into llama-quant-refactor
87be6a36
Merge branch 'ggml-org:master' into llama-quant-refactor
9c00aab2
Manually set proper ordering of tensors, mostly applies to gemma
04c82996
Add quantization recipes
d51eb1d1
Fix compile warnings
072ac799
Update tests
263bf232
Add specific attention v counters and a way to group them like in leg…
d7ceecb0
Reviewers
No reviews
Assignees
No one assigned
Labels
testing
examples
Milestone
No milestone
Login to write a write a comment.
Login via GitHub