Commit History

CUDA: backwards pass for misc. ops, add tests (llama/11257)
2fbcec1

JohannesGaessler commited on

ggml: aarch64: implement SVE kernels for q4_K_q8_K vector dot (llama/11227)
bf3dc93

fj-y-saito ggerganov commited on

vulkan: scale caching for k quants + misc fixes (llama/11081)
03ab36f

Eve commited on

fix: ggml: fix vulkan-shaders-gen build (llama/10448)
ad8f031

Sparkleholic commited on

RoPE: fix back, CUDA support for back + noncont. (llama/11240)
131a21e

JohannesGaessler commited on

SYCL: Add gated linear attention kernel (llama/11175)
fdb1fe5

qnixsynapse commited on

ggml : add option to not print stack on abort (ggml/1081)
9b2706e

William Tambellini Diego Devesa commited on

ggml-cpu : fix ggml_graph_compute_thread did not terminate on abort. (ggml/1065)
8e57313

issixx issi commited on

ci : dummy commit to trigger CI
600a548
unverified

ggerganov commited on

ruby : Make context accept initial parameters, API to retrieve a segment and more (#2749)
7cb9a0e
unverified

KitaitiMakoto commited on

whisper.objc : fix build and CI
9cbd99a
unverified

Corey Earwood commited on

talk-llama : sync llama.cpp
16d40d7

ggerganov commited on

sync : ggml
d50f71a

ggerganov commited on

GGUF: C++ refactor, backend support, misc fixes (skip) (llama/11030)
92311a3

JohannesGaessler commited on

ggml : add opencl backend (skip) (llama/10693)
226358f

lhez Skyler Szot Shangqing Gu Alexander Angus Hongqiang Wang Max Krasnyansky commited on

cuda : CUDA Graph Compute Function Refactor (precursor for performance improvements) (llama/11042)
25882f6

Andreas Kieslinger slaren commited on

ggml : do not define GGML_USE_CUDA when building with GGML_BACKEND_DL (llama/11211)
79f750d

rgerganov commited on

Vulkan: Fix float16 use on devices without float16 support + fix subgroup_size_control validation error (llama/11161)
5ad3f1d

OccamRazor commited on

SYCL: Refactor ggml_sycl_compute_forward (llama/11121)
fa23a38

qnixsynapse commited on

fix: add missing msg in static_assert (llama/11143)
8c60d6a

hydaitw commited on

llamafile : ppc64le MMA INT8 implementation (llama/10912)
6f18eed

amritahs-ibm commited on

Disable GL_KHR_cooperative_matrix Vulkan extension if not available. (llama/11117)
623b74d

mbaudier commited on

fix: Vulkan shader gen binary path when Cross-compiling (llama/11096)
966a7bb

ag2s20150909 commited on

GGUF: C++ refactor, backend support, misc fixes (llama/11030)
21c5b64

JohannesGaessler commited on

ggml-backend : only offload from host buffers (fix) (llama/11124)
9ac3c7e

Diego Devesa commited on

ggml-backend : only offload from host buffers (llama/11120)
1ca87a8

Diego Devesa commited on

rpc : code cleanup (llama/11107)
a0fb22d

rgerganov commited on

SYCL: Use get_multi_ptr instead of deprecated get_pointer in wkv6 (llama/11087)
4ed93cc

qnixsynapse commited on

CUDA: add BF16 support (llama/11093)
961ef57

JohannesGaessler commited on

Vulkan: Add device-specific blacklist for coopmat for the AMD proprietary driver (llama/11074)
4d90c3d

OccamRazor commited on

Support for models with non-512-aligned tensors over RPC. (llama/11047)
895a3a2

Billy462 Diego Devesa commited on

fix: Vulkan shader gen binary path (llama/11037)
7008fb8

Gilad S. commited on

ggml : allow loading backend with env variable (ggml/1059)
48aa6d0

rgerganov commited on

scripts : sync opencl, gguf
f751550
unverified

ggerganov commited on

whisper : fix gpu device selection (#2728)
87b427e
unverified

ggerganov commited on

server : fix build (#2718)
7925ae3
unverified

ggerganov commited on

talk-llama : sync llama.cpp (#2709)
b462700
unverified

ggerganov commited on

server : generate unique tmp filenames (#2718)
89d94b1
unverified

NETZkultur GmbH commited on

whisper : add whisper_full_get_segment_no_speech_prob_from_state (#2716)
cb32a92
unverified

Sandro Hanea commited on

readme : add docker instructions (#2711)
28257a6
unverified

jayant-yadav commited on

docs: Fix main -> whisper-cli in download scripts (#2707)
4abfe5a
unverified

Adam Jones commited on

release : v1.7.4
c775ca4
unverified

ggerganov commited on

ci : cont
6331634
unverified

ggerganov commited on

ci : fix ubuntu runner names
9a3c061
unverified

ggerganov commited on

cli : fix segfault on missing argument (#2700)
245a91f
unverified

Yusuf Redžić commited on

ci : fix arm builds
31f91d9

ggerganov commited on

sync : ggml
0211dda

ggerganov commited on

ggml : do not install metal source when embed library (ggml/1054)
9615cf2

ggerganov commited on

metal : avoid uint (llama/11019)
b788516

ggerganov commited on