 Johannes Gäßler
|
affc76edfd
cuda : loading models directly into VRAM, norm calculation on GPU, broadcasting for ggml_mul (#1483)
|
2 лет назад |
 Georgi Gerganov
|
ea600071cb
Revert "feature : add blis and other BLAS implementation support (#1502)"
|
2 лет назад |
 Zenix
|
07e9ace0f9
feature : add blis and other BLAS implementation support (#1502)
|
2 лет назад |
|
Georgi Gerganov
|
ec2e10c444
llama : add llama_init_backend() API (close #1527)
|
2 лет назад |
 DannyDaemonic
|
d2c59b8ba4
Fix for mingw (#1462)
|
2 лет назад |
 Maxime
|
503db28849
llama : fix name shadowing and C4146 (#1526)
|
2 лет назад |
|
Georgi Gerganov
|
8a203f9fa1
llama : fix compile warnings in llama_set_state_data()
|
2 лет назад |
|
Georgi Gerganov
|
4fd3e29297
ggml : fix scalar implementation of Q4_1 dot
|
2 лет назад |
|
Georgi Gerganov
|
2d5db48371
ggml : use F16 instead of F32 in Q4_0, Q4_1, Q8_0 (#1508)
|
2 лет назад |
|
Georgi Gerganov
|
6986c7835a
tests : add missing header
|
2 лет назад |
 Evan Jones
|
943e6081cc
examples : add persistent chat (#1495)
|
2 лет назад |
 Jason McCartney
|
7694b52b9a
main : make reverse prompt option act as a stop token in non-interactive mode (#1032)
|
2 лет назад |
 David Kennedy
|
79e3efb0e9
readme : adds WizardLM to the list of supported models (#1485)
|
2 лет назад |
|
Georgi Gerganov
|
4b7e245adf
minor : fix compile warnings
|
2 лет назад |
 Erik Scholz
|
5ea4339273
make kv_f16 the default for api users (#1517)
|
2 лет назад |
|
DannyDaemonic
|
ee9654138a
Fixes #1511 lambda issue for w64devkit (mingw) (#1513)
|
2 лет назад |
 Stephan Walter
|
dc271c52ed
Remove unused n_parts parameter (#1509)
|
2 лет назад |
 rankaiyx
|
c238b5873a
benchmark-matmul: Print the average of the test results (#1490)
|
2 лет назад |
 Tom Jobbins
|
2b2646931b
convert.py: Support models which are stored in a single pytorch_model.bin (#1469)
|
2 лет назад |
 Ilya Kurdyukov
|
42627421ec
~7% faster Q5_1 AVX2 code (#1477)
|
2 лет назад |
 András Salamon
|
9560655409
define default model path once, sync path with readme (#1366)
|
2 лет назад |
 sandyiscool
|
2a5ee023ad
Add alternate include path for openblas (#1476)
|
2 лет назад |
 zrm
|
63d20469b8
fix get_num_physical_cores() (#1436)
|
2 лет назад |
 slaren
|
b5c9295eef
benchmark-matmul: fix clang-tidy issues, report results in GFLOPS (#1458)
|
2 лет назад |
|
Johannes Gäßler
|
eb363627fd
cuda : deduplicated dequantization code (#1453)
|
2 лет назад |
 xaedes
|
79b2d5b69d
ggml : alternative fix for race condition bug in non-inplace ggml_compute_forward_diag_mask_f32 (#1454)
|
2 лет назад |
|
Georgi Gerganov
|
13c351ad72
ggml : various fixes (#1450)
|
2 лет назад |
 katsu560
|
60f8c361ca
ggml : add AVX support based on AVX2 code (#1430)
|
2 лет назад |
|
Georgi Gerganov
|
601a033475
ggml : add GGML_QNT_VERSION to track quantization format changes
|
2 лет назад |
|
Georgi Gerganov
|
08737ef720
cuda : fix convert function (#1412)
|
2 лет назад |
