 Carolinabanana
|
5dc9dd7152
llama : add Command R Plus support (#6491)
|
vor 1 Jahr |
 Kawrakow
|
55c1b2a3bb
IQ1_M: 1.75 bpw quantization (#6302)
|
vor 1 Jahr |
 Georgi Gerganov
|
8030da7afe
ggml : reuse quantum structs across backends (#5943)
|
vor 1 Jahr |
|
Kawrakow
|
be858f6205
Better 1.5 bit quantization (#5971)
|
vor 1 Jahr |
|
Georgi Gerganov
|
5b09797321
ggml : remove old quantization functions (#5942)
|
vor 1 Jahr |
|
Georgi Gerganov
|
8a3012a4ad
ggml : add ggml-common.h to deduplicate shared code (#5940)
|
vor 1 Jahr |
|
Kawrakow
|
7c4263d426
ggml : make i-quants work with super-blocks of 64 (CPU,Metal) (#5760)
|
vor 1 Jahr |
|
Kawrakow
|
0becb22ac0
IQ4_XS: a 4.25 bpw quantization (#5747)
|
vor 1 Jahr |
|
Kawrakow
|
a33e6a0d2a
Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (#5721)
|
vor 1 Jahr |
|
Kawrakow
|
4c4cb30736
IQ3_S: a much better alternative to Q3_K (#5676)
|
vor 1 Jahr |
|
Kawrakow
|
a14679cc30
IQ4_NL: 4-bit non-linear quants with blocks of 32 (#5590)
|
vor 1 Jahr |
|
Kawrakow
|
bd2d4e393b
1.5 bit quantization (#5453)
|
vor 1 Jahr |
 snadampal
|
a07d0fee1f
ggml : add mmla kernels for quantized GEMM (#4966)
|
vor 1 Jahr |
|
Kawrakow
|
c6b395535a
ggml : make use of ggml-quants.h possible in C++ code (#5338)
|
vor 1 Jahr |
|
Kawrakow
|
f4d7e54974
SOTA 3-bit quants (#5196)
|
vor 1 Jahr |
|
Georgi Gerganov
|
38566680cd
ggml : add IQ2 to test-backend-ops + refactoring (#4990)
|
vor 2 Jahren |
|
Kawrakow
|
334a835a1c
ggml : importance matrix support for legacy quants (#4969)
|
vor 2 Jahren |
|
Kawrakow
|
467a882fd2
Add ability to use importance matrix for all k-quants (#4930)
|
vor 2 Jahren |
|
Kawrakow
|
147b17ac94
2-bit quantizations (#4897)
|
vor 2 Jahren |
|
Kawrakow
|
49662cbed3
ggml : SOTA 2-bit quants (add IQ2_XS) (#4856)
|
vor 2 Jahren |
|
Kawrakow
|
dd5ae06405
SOTA 2-bit quants (#4773)
|
vor 2 Jahren |
|
Georgi Gerganov
|
d061bf9405
ggml : fix q2_k bpw in comments (ggml/680)
|
vor 2 Jahren |
|
Georgi Gerganov
|
207b51900e
ggml : move FP16 <-> FP32 code to ggml-impl.h (#3861)
|
vor 2 Jahren |
|
Georgi Gerganov
|
d69d777c02
ggml : quantization refactoring (#3833)
|
vor 2 Jahren |
