slaren
|
2d08b7fbb4
docker : build only main and server in their images (#7782)
|
преди 1 година |
slaren
|
d67caea0d6
docker : add openmp lib (#7780)
|
преди 1 година |
Galunid
|
7672adeec7
Fix encoding in python scripts (#7733)
|
преди 1 година |
Johannes Gäßler
|
7d1a378b8f
CUDA: refactor mmq, dmmv, mmvq (#7716)
|
преди 1 година |
Georgi Gerganov
|
2b3389677a
ggml : refactor rope norm/neox (#7634)
|
преди 1 година |
arch-btw
|
9973e81c5c
readme : remove -ins (#7759)
|
преди 1 година |
jaime-m-p
|
c90dbe026b
Fix per token atrributes bits (#7749)
|
преди 1 година |
agray3
|
b90dc566c1
Allow number of nodes in CUDA graph to change (#7738)
|
преди 1 година |
Georgi Gerganov
|
1442677f92
common : refactor cli arg parsing (#7675)
|
преди 1 година |
Georgi Gerganov
|
554c247caf
ggml : remove OpenCL (#7735)
|
преди 1 година |
Georgi Gerganov
|
0cd6bd3483
llama : remove beam search (#7736)
|
преди 1 година |
Georgi Gerganov
|
5ca0944a15
readme : remove obsolete Zig instructions (#7471)
|
преди 1 година |
slaren
|
adc9ff3841
llama-bench : allow using a different printer for stderr with -oe (#7722)
|
преди 1 година |
Daniele
|
987d743d6b
Improve hipBLAS support in CMake (#7696)
|
преди 1 година |
zhouwg
|
b226c1227b
refine .gitignore (#7688)
|
преди 1 година |
jaime-m-p
|
3b38d48609
Per token attributes (#7685)
|
преди 1 година |
Georgi Gerganov
|
6d1616944d
ggml : prevent builds with -ffinite-math-only (#7726)
|
преди 1 година |
Radoslav Gerganov
|
bde7cd3cd9
llama : offload to RPC in addition to other backends (#7640)
|
преди 1 година |
Masaya, Kato
|
a5735e4426
ggml : use OpenMP as a thread pool (#7606)
|
преди 1 година |
Johannes Gäßler
|
0b832d53ba
make: fix debug options not being applied to NVCC (#7714)
|
преди 1 година |
0cc4m
|
3d7ebf6312
Vulkan Mixture of Experts (MoE) support (#7628)
|
преди 1 година |
Andy Tai
|
a10cda58d3
cmake : add pkg-config spec file for llama.cpp (#7702)
|
преди 1 година |
zhangkaihuo
|
6f28a333c1
llama : MiniCPM support tied embeddings (#7664)
|
преди 1 година |
Georgi Gerganov
|
549279d804
llama : avoid double token-to-piece cache (#7654)
|
преди 1 година |
woachk
|
9e405b6e2e
kompute : implement op_getrows_f32 (#6403)
|
преди 1 година |
Dave Airlie
|
3413ae2193
fix bug introduced in using calloc (#7701)
|
преди 1 година |
Georgi Gerganov
|
1669810d7c
flake.lock: Update (#7686)
|
преди 1 година |
Austin
|
7c4e5b7eae
chore : add ignore rule for generated server themes (#7689)
|
преди 1 година |
nickp27
|
9422c5e34b
[SYCL] Update rpc-server.cpp to include SYCL backend (#7682)
|
преди 1 година |
Johannes Gäßler
|
e141ce624a
Fix FlashAttention debug test, FP32 assert (#7684)
|
преди 1 година |