 slaren
|
da0400344b
ggml-cuda : perform cublas fp16 matrix multiplication as fp16 (#3370)
|
2 lat temu |
 Zhang Peiyuan
|
e519621010
convert : remove bug in convert.py permute function (#3364)
|
2 lat temu |
 Richard Roberson
|
ac43576124
make-ggml.py : compatibility with more models and GGUF (#3290)
|
2 lat temu |
 Cebtenzzre
|
20c7e1e804
gguf : fix a few general keys (#3341)
|
2 lat temu |
 Rickard Hallerbäck
|
dc6897404e
metal : reusing llama.cpp logging (#3152)
|
2 lat temu |
 Jag Chadha
|
527e57cfd8
build : add ACCELERATE_NEW_LAPACK to fix warning on macOS Sonoma (#3342)
|
2 lat temu |
 BarfingLemurs
|
ffe88a36a9
readme : add some recent perplexity and bpw measurements to READMES, link for k-quants (#3340)
|
2 lat temu |
 DAN™
|
99115f3fa6
cmake : fix build-info.h on MSVC (#3309)
|
2 lat temu |
 2f38b454
|
1726f9626f
docs: Fix typo CLBlast_DIR var. (#3330)
|
2 lat temu |
 Erik Scholz
|
a98b1633d5
nix : add cuda, use a symlinked toolkit for cmake (#3202)
|
2 lat temu |
|
slaren
|
c091cdfb24
llama-bench : add README (#3317)
|
2 lat temu |
|
Cebtenzzre
|
51a7cf5c6e
examples : fix RoPE defaults to match PR #3240 (#3315)
|
2 lat temu |
 Kevin Ji
|
bedb92b603
scripts : use `/usr/bin/env` in shebang (#3313)
|
2 lat temu |
 Lee Drake
|
bc9d3e3971
Update README.md (#3289)
|
2 lat temu |
 shibe2
|
36b904e200
ggml-opencl.cpp: Make private functions static (#3300)
|
2 lat temu |
 Edward Taylor
|
324f3403d5
zig : fix for updated c lib (#3259)
|
2 lat temu |
 yuiseki
|
f56c418ab0
embedding : update README.md (#3224)
|
2 lat temu |
 Johannes Gäßler
|
8185710a80
CUDA: use only 1 thread if fully offloaded (#2915)
|
2 lat temu |
 Georgi Gerganov
|
7eb41179ed
readme : update hot topics
|
2 lat temu |
|
Cebtenzzre
|
a5661d7e71
llama : allow gguf RoPE keys to be overridden with defaults (#3240)
|
2 lat temu |
|
Cebtenzzre
|
65c2c1c5ab
benchmark-matmult : do not use integer abs() on a float (#3277)
|
2 lat temu |
 kang
|
80834daecf
flake : Restore default package's buildInputs (#3262)
|
2 lat temu |
 Alon
|
a40f2b656f
CI: FreeBSD fix (#3258)
|
2 lat temu |
|
Georgi Gerganov
|
d119c04c15
examples : fix benchmark-matmult (#1554)
|
2 lat temu |
|
Cebtenzzre
|
8781013ef6
make : restore build-info.h dependency for several targets (#3205)
|
2 lat temu |
|
Erik Scholz
|
7ddf185537
ci : switch cudatoolkit install on windows to networked (#3236)
|
2 lat temu |
|
Johannes Gäßler
|
ee66942d7e
CUDA: fix peer access logic (#3231)
|
2 lat temu |
|
Johannes Gäßler
|
111163e246
CUDA: enable peer access between devices (#2470)
|
2 lat temu |
|
slaren
|
8b428c9bc8
llama.cpp : show model size and BPW on load (#3223)
|
2 lat temu |
|
Johannes Gäßler
|
578d8c8f5c
CUDA: fix scratch malloced on non-main device (#3220)
|
2 lat temu |
