llama.cpp
quantize: Handle user-defined quantization levels for additional tensors
#12511
Merged

quantize: Handle user-defined quantization levels for additional tensors #12511

ggerganov merged 35 commits into ggml-org:master from EAddario:quantize
EAddario
EAddario Add llama_model_quantize_params parameters
09f716d7
EAddario Add new quantize parameters parsing and validation
ac908af2
EAddario Update usage
337d9792
EAddario Add new parameters defaults
6f8d16dc
EAddario Add new quantization parameters logic
71c9f93e
EAddario Add llama_model_quantize_params parameters
8e18131b
EAddario Add new quantize parameters parsing and validation
a77d9470
EAddario Update usage
2414eaa9
EAddario Add new parameters defaults
0dd66b81
EAddario Add new quantization parameters logic
1d841c67
EAddario Merge main changes into branch
120f71b7
EAddario Merge branch 'master' into quantize
dbcc0b5a
EAddario Minor refactoring as per the contributors' coding guidelines
d86de03c
EAddario Update descriptions to match existing style
99bae5e9
EAddario Merge branch 'master' into quantize
60b0a53e
EAddario Merge branch 'master' into quantize
3e2063d4
EAddario Merge branch 'master' into quantize
b99fa62b
EAddario Add llama_model_quantize_params parameters
f97b693a
EAddario Add new quantize parameters parsing and validation
f11e3da2
EAddario Update usage
ad1e3524
EAddario Add new parameters defaults
4e5c96a3
EAddario Add new quantization parameters logic
9b3ccb53
EAddario Minor refactoring as per the contributors' guidelines
35f45f19
EAddario Merge branch 'master' into quantize
071e9ef2
github-actions github-actions added examples
EAddario EAddario changed the title Handle user-defined quantization levels for additional tensors quantize: Handle user-defined quantization levels for additional tensors 203 days ago
max-krasnyansky
EAddario
jukofyork
EAddario Implement general --tensor-type instead of tensor-specific command op…
54e13cf6
EAddario Merge branch 'master' into quantize
31d642c5
EAddario
ddh0
ddh0
EAddario Fix implied type bug
b3c7db57
EAddario
joseph777111
ddh0
ddh0
EAddario Restore missing #includes
625f0ae5
EAddario
EAddario Add regex capability for tensor selection
2fd0b41f
EAddario Merge branch 'master' into quantize
3e9f5658
EAddario
jukofyork
ngxson
ngxson commented on 2025-04-02
ddh0
EAddario Refactor function name and update ALLOWED_TENSOR_TYPE
054ede4e
EAddario Add missing #include
5a304b8e
ddh0
EAddario Handle edge case when tensor name is cls.output
1acb9f4a
EAddario
ubergarm
David-AU-github
EAddario
EAddario Minor logging improvement
04604a46
EAddario Merge branch 'master' into quantize
30443a5b
David-AU-github
ggerganov
ggerganov approved these changes on 2025-04-08
ggerganov ggerganov requested a review from slaren slaren 183 days ago
slaren
slaren approved these changes on 2025-04-11
EAddario
EAddario
slaren
EAddario
ggerganov ggerganov merged 71e90e88 into master 180 days ago
acbits
EAddario
EAddario EAddario deleted the quantize branch 180 days ago
joseph777111
Djip007
EAddario
Djip007
EAddario

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone