 Kawrakow
|
bd2d4e393b
1.5 bit quantization (#5453)
|
1 year ago |
|
Kawrakow
|
895407f31b
ggml-quants : fix compiler warnings (shadow variable) (#5472)
|
1 year ago |
 Georgi Gerganov
|
0f2411f154
ggml : fix compile warnings (unused vars) (#4966)
|
1 year ago |
 snadampal
|
a07d0fee1f
ggml : add mmla kernels for quantized GEMM (#4966)
|
1 year ago |
 Michael Podvitskiy
|
b2f87cb64d
ggml : fix `error C2078: too many initializers` for MSVC ARM64 (#5404)
|
2 years ago |
|
Kawrakow
|
f57fadc009
Slight quantization improvement for Q4_K and Q5_K (#5361)
|
2 years ago |
|
Kawrakow
|
6fdfa2ecc6
iq2_xxs: tune quantization (#5320)
|
2 years ago |
|
Kawrakow
|
8e14e3ddb3
Faster AVX2 dot product for IQ2_XS (#5187)
|
2 years ago |
|
Kawrakow
|
f4d7e54974
SOTA 3-bit quants (#5196)
|
2 years ago |
|
Georgi Gerganov
|
38566680cd
ggml : add IQ2 to test-backend-ops + refactoring (#4990)
|
2 years ago |
|
Kawrakow
|
334a835a1c
ggml : importance matrix support for legacy quants (#4969)
|
2 years ago |
|
Kawrakow
|
467a882fd2
Add ability to use importance matrix for all k-quants (#4930)
|
2 years ago |
|
Kawrakow
|
147b17ac94
2-bit quantizations (#4897)
|
2 years ago |
|
Georgi Gerganov
|
f238461236
ggml : fix 32-bit ARM compat for IQ2_XS (whisper/1758)
|
2 years ago |
|
Kawrakow
|
49662cbed3
ggml : SOTA 2-bit quants (add IQ2_XS) (#4856)
|
2 years ago |
|
Georgi Gerganov
|
18c2e1752c
ggml : fix vld1q_s8_x4 32-bit compat (#4828)
|
2 years ago |
|
Kawrakow
|
dd5ae06405
SOTA 2-bit quants (#4773)
|
2 years ago |
|
Georgi Gerganov
|
e39106c055
ggml : add ggml_vdotq_s32 alias (#4715)
|
2 years ago |
|
Georgi Gerganov
|
951010fa53
ggml : fix dot product for ARM (#4630)
|
2 years ago |
 FantasyGmm
|
a55876955b
cuda : fix jetson compile error (#4560)
|
2 years ago |
 Richard Kiss
|
9494d7c477
english : use `typos` to fix comments and logs (#4354)
|
2 years ago |
 Roger Meier
|
8e9361089d
build : support ppc64le build for make and CMake (#3963)
|
2 years ago |
 Michael Potter
|
6bb4908a17
Fix MacOS Sonoma model quantization (#4052)
|
2 years ago |
|
Georgi Gerganov
|
3d68f364f1
ggml : sync (im2col, GPU conv, 32-bit arm compat) (#4060)
|
2 years ago |
|
Georgi Gerganov
|
9a3b4f6c86
ggml : fix UNUSED macro (#3762)
|
2 years ago |
 Andrew Godfrey
|
73bdcb395e
finetune : add -ngl parameter (#3762)
|
2 years ago |
|
Georgi Gerganov
|
207b51900e
ggml : move FP16 <-> FP32 code to ggml-impl.h (#3861)
|
2 years ago |
|
Georgi Gerganov
|
d69d777c02
ggml : quantization refactoring (#3833)
|
2 years ago |
