 Johannes Gäßler
|
5e116e8dd5
make/cmake: add missing force MMQ/cuBLAS for HIP (#8515)
|
il y a 1 an |
 bandoti
|
17eb6aa8a9
vulkan : cmake integration (#8119)
|
il y a 1 an |
 Nicholai Tukanov
|
368645698a
ggml : add NVPL BLAS support (#8329) (#8425)
|
il y a 1 an |
 Clint Herron
|
dd07a123b7
Name Migration: Build the deprecation-warning 'main' binary every time (#8404)
|
il y a 1 an |
 Georgi Gerganov
|
6b2a849d1f
ggml : move sgemm sources to llamafile subfolder (#8394)
|
il y a 1 an |
 Dibakar Gope
|
0f1a39f343
ggml : add AArch64 optimized GEMV and GEMM Q4 kernels (#5780)
|
il y a 1 an |
|
Clint Herron
|
e500d6135a
Deprecation warning to assist with migration to new binary names (#8283)
|
il y a 1 an |
|
Johannes Gäßler
|
a03e8dd99d
make/cmake: LLAMA_NO_CCACHE -> GGML_NO_CCACHE (#8392)
|
il y a 1 an |
 Brian
|
f7cab35ef9
gguf-hash: model wide and per tensor hashing using xxhash and sha1 (#8048)
|
il y a 1 an |
|
Clint Herron
|
3e2618bc7b
Adding step to `clean` target to remove legacy binary names to reduce upgrade / migration confusion arising from #7809. (#8257)
|
il y a 1 an |
 Xuan Son Nguyen
|
a27aa50ab7
Add missing items in makefile (#8177)
|
il y a 1 an |
 slaren
|
c7ab7b612c
make : fix missing -O3 (#8143)
|
il y a 1 an |
|
Georgi Gerganov
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
il y a 1 an |
|
Johannes Gäßler
|
a818f3028d
CUDA: use MMQ instead of cuBLAS by default (#8075)
|
il y a 1 an |
|
slaren
|
95f57bb5d5
ggml : remove ggml_task_type and GGML_PERF (#8017)
|
il y a 1 an |
|
Clint Herron
|
c5a8d4b749
JSON Schema to GBNF integration tests (#7790)
|
il y a 1 an |
 Ulrich Drepper
|
61665277af
Allow compiling with CUDA without CUDA runtime installed (#7989)
|
il y a 1 an |
 0cc4m
|
7c7836d9d4
Vulkan Shader Refactor, Memory Debugging Option (#7947)
|
il y a 1 an |
|
Xuan Son Nguyen
|
0c7b3595b9
Add `cvector-generator` example (#7514)
|
il y a 1 an |
|
slaren
|
f578b86b21
move BLAS to a separate backend (#6210)
|
il y a 1 an |
 Olivier Chafik
|
1c641e6aac
`build`: rename main → llama-cli, server → llama-server, llava-cli → llama-llava-cli, etc... (#7809)
|
il y a 1 an |
|
Johannes Gäßler
|
7d1a378b8f
CUDA: refactor mmq, dmmv, mmvq (#7716)
|
il y a 1 an |
|
Georgi Gerganov
|
554c247caf
ggml : remove OpenCL (#7735)
|
il y a 1 an |
|
Georgi Gerganov
|
0cd6bd3483
llama : remove beam search (#7736)
|
il y a 1 an |
 Radoslav Gerganov
|
bde7cd3cd9
llama : offload to RPC in addition to other backends (#7640)
|
il y a 1 an |
 Masaya, Kato
|
a5735e4426
ggml : use OpenMP as a thread pool (#7606)
|
il y a 1 an |
|
Johannes Gäßler
|
0b832d53ba
make: fix debug options not being applied to NVCC (#7714)
|
il y a 1 an |
 Yazan Agha-Schrader
|
2e666832e6
server : new UI (#7633)
|
il y a 1 an |
|
Johannes Gäßler
|
9b596417af
CUDA: quantized KV support for FA vec (#7527)
|
il y a 1 an |
 Daniele
|
30e238b246
Improve HIP compatibility (#7672)
|
il y a 1 an |
