Georgi Gerganov
|
e128a1bf5b
tests : fix test-quantize-fns to init the CPU backend (#12306)
|
10 mesiacov pred |
Georgi Gerganov
|
f6d12e7df8
tests : fix compile warning
|
1 rok pred |
Diego Devesa
|
ae8de6d50a
ggml : build backends as libraries (#10256)
|
1 rok pred |
Diego Devesa
|
9f40989351
ggml : move CPU backend to a separate file (#10144)
|
1 rok pred |
Diego Devesa
|
dca1d4b58a
ggml : fix BLAS with unsupported types (#9775)
|
1 rok pred |
compilade
|
9bc6db28d0
ggml-quants : ternary packing for TriLMs and BitNet b1.58 (#8151)
|
1 rok pred |
Georgi Gerganov
|
370b1f7e7a
ggml : minor naming changes (#8433)
|
1 rok pred |
Kawrakow
|
1f2fd4e727
tests : include IQ2_XXS and IQ2_XS in test-quantize-fns (#6303)
|
1 rok pred |
Kawrakow
|
a33e6a0d2a
Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (#5721)
|
1 rok pred |
Kawrakow
|
4c4cb30736
IQ3_S: a much better alternative to Q3_K (#5676)
|
1 rok pred |
snadampal
|
a07d0fee1f
ggml : add mmla kernels for quantized GEMM (#4966)
|
1 rok pred |
Kawrakow
|
f4d7e54974
SOTA 3-bit quants (#5196)
|
1 rok pred |
Kawrakow
|
49662cbed3
ggml : SOTA 2-bit quants (add IQ2_XS) (#4856)
|
2 rokov pred |
Kawrakow
|
dd5ae06405
SOTA 2-bit quants (#4773)
|
2 rokov pred |
Georgi Gerganov
|
207b51900e
ggml : move FP16 <-> FP32 code to ggml-impl.h (#3861)
|
2 rokov pred |
Cebtenzzre
|
3aefaab9e5
check C++ code with -Wmissing-declarations (#3184)
|
2 rokov pred |
Stephan Walter
|
1b107b8550
ggml : generalize `quantize_fns` for simpler FP16 handling (#1237)
|
2 rokov pred |
Borislav Stanimirov
|
9cbf50c041
build : fix and ignore MSVC warnings (#1889)
|
2 rokov pred |
Kawrakow
|
99009e72f8
ggml : add SOTA 2,3,4,5,6 bit k-quantizations (#1684)
|
2 rokov pred |
Georgi Gerganov
|
7a32fcb3b2
ggml : add Q8_0 quantization format (rename the old one to Q8_1) (ARM NEON) (#1179)
|
2 rokov pred |
Stephan Walter
|
c50b628810
Fix CI: ARM NEON, quantization unit tests, editorconfig (#1122)
|
2 rokov pred |
unbounded
|
5f939498d5
ggml : unit test for quantization functions (#953)
|
2 rokov pred |