 slaren
|
ebc96086af
ggml-alloc : correctly check mmap return value for errors (#3075)
|
2 жил өмнө |
 Kunshang Ji
|
7f412dab9c
enable CPU HBM (#2603)
|
2 жил өмнө |
 Cebtenzzre
|
6336d834ec
convert : fix F32 ftype not being saved (#3048)
|
2 жил өмнө |
|
Cebtenzzre
|
00d62adb79
fix some warnings from gcc and clang-tidy (#3038)
|
2 жил өмнө |
|
Cebtenzzre
|
4fa2cc1750
make : improve test target (#3031)
|
2 жил өмнө |
|
Cebtenzzre
|
5ffab089a5
make : fix CPPFLAGS (#3035)
|
2 жил өмнө |
|
slaren
|
15b67a66c2
llama-bench : use two tokens in the warmup run for prompt evals (#3059)
|
2 жил өмнө |
 Kawrakow
|
be8c9c245b
metal : parallel RoPE on Metal (#3024)
|
2 жил өмнө |
|
Kawrakow
|
be6beeb8d7
metal : correct fix of kernel_norm (#3060)
|
2 жил өмнө |
 Georgi Gerganov
|
c4f496648c
metal : fix kernel_norm (fixes Falcon on Metal) (#3057)
|
2 жил өмнө |
 Przemysław Pawełczyk
|
fec2fb19e4
ggml : posixify madvise and pagesize (#3037)
|
2 жил өмнө |
|
Georgi Gerganov
|
178b1850eb
k-quants : fix zero-weight guard in Q6_K (ref #3040)
|
2 жил өмнө |
 Kerfuffle
|
ea2c85d5d2
convert-llama-ggml-to-gguf: Try to handle files older than GGJTv3 (#3023)
|
2 жил өмнө |
|
Cebtenzzre
|
9912b9efc8
build : add LLAMA_METAL_NDEBUG flag (#3033)
|
2 жил өмнө |
|
Cebtenzzre
|
9e2023156e
make : use new flag variables for recent changes (#3019)
|
2 жил өмнө |
|
Cebtenzzre
|
de2fe892af
examples : replace fprintf to stdout with printf (#3017)
|
2 жил өмнө |
 Erik Scholz
|
c9c3220c48
convert: fix convert.py not working with int filename_stem (#3028)
|
2 жил өмнө |
|
Kawrakow
|
d59bd97065
Guard against all weights in a super-block being zero (#3010)
|
2 жил өмнө |
|
Georgi Gerganov
|
35938ee3b0
llama : update logic for number of threads when using BLAS
|
2 жил өмнө |
|
Georgi Gerganov
|
921772104b
speculative : add grammar support (#2991)
|
2 жил өмнө |
|
Georgi Gerganov
|
2ba85c8609
py : minor
|
2 жил өмнө |
|
Georgi Gerganov
|
e36ecdccc8
build : on Mac OS enable Metal by default (#2901)
|
2 жил өмнө |
|
slaren
|
bd33e5ab92
ggml-opencl : store GPU buffer in ggml_tensor::extra (#2994)
|
2 жил өмнө |
|
Cebtenzzre
|
3103568144
llama-bench : make cpp file non-executable (#2999)
|
2 жил өмнө |
 Leng Yue
|
5b8530d88c
make : add speculative example (#3003)
|
2 жил өмнө |
 Aarni Koskela
|
e4386f417f
server : add a subtle loading animation to the edit box (#2466)
|
2 жил өмнө |
 Jiahao Li
|
35195689cd
2x faster (rms) norm cuda kernels (3.7% e2e improvement) (#2985)
|
2 жил өмнө |
|
slaren
|
cf9b08485c
ggml-alloc : use virtual memory for measurement (#2973)
|
2 жил өмнө |
|
Georgi Gerganov
|
47068e5170
speculative : PoC for speeding-up inference via speculative sampling (#2926)
|
2 жил өмнө |
|
Georgi Gerganov
|
8f429fa511
perplexity : fix ETA by warming up the model with an empty run
|
2 жил өмнө |
