Commit History

ggml : add error handling to graph_compute (#1714)
92f24ee
unverified

finnvoorhees commited on

cuda : simplify expression
cda4a91

ggerganov slaren commited on

cuda : mark I16 and I32 ops as unsupported
cec288d

ggerganov commited on

metal : add kernel_get_rows_i32
459dd87

ggerganov commited on

metal : optimize ggml_mul_mat_id (faster Mixtral PP) (llama/4725)
8bc6274

ggerganov commited on

metal : enable shader debugging (cmake option) (llama/4705)
7dd37dc

ggerganov commited on

ggml : add ggml_vdotq_s32 alias (llama/4715)
96dc902

ggerganov commited on

CUDA: fixed tensor cores not being used on RDNA3 (llama/4697)
654d245

JohannesGaessler commited on

ggml : add ggml_cpu_has_avx_vnni() (llama/4589)
b10cbfd

alandao ggerganov commited on

CUDA: fix tensor core logic for Pascal and HIP (llama/4682)
977baeb

JohannesGaessler commited on

cuda: fix vmm oom issue on NVIDIA AGX Orin (llama/4687)
6980ee4

hydaitw commited on

ggml : extend ggml_get_rows, ggml_repeat, ggml_concat (ggml/639)
f17d170

Guillaume Wenzek ggerganov commited on

scripts : fix sync order + metal sed
1785412

ggerganov commited on

examples : fix WASM Stack Overflow (#1713)
57c0e54
unverified

AHuguet commited on

docker : fix the publishing of the CUDA Docker image (#1704)
6091193
unverified

bobqianic commited on

scripts : do not sync commits from this repo
397f291
unverified

ggerganov commited on

ci : build with CLBlast + ggml-opencl use GGML_API (#1576)
41a13d4
unverified

Tamotsu Takahashi commited on

whisper : replace `tensor->n_dims` with `ggml_n_dims(tensor)` (#1694)
cee2822
unverified

bobqianic commited on

sync : ggml (VMM, sync-ggml-am, dotprod ARM fixes, CUDA fixes) (#1691)
919a447
unverified

ggerganov commited on

download : fix large q5 model name (#1695)
5df6c6c
unverified

DimoP commited on

whisper : Replace WHISPER_PRINT_DEBUG with WHISPER_LOG_DEBUG (#1681)
5ad04c9
unverified

bobqianic commited on

sync : ggml (ggml_scale, ggml_row_size, etc.) (#1677)
aa86ade
unverified

ggerganov commited on

docker : Dockerize whisper.cpp (#1674)
7163150
unverified

Chaoqun commited on

CI : Add coverage for talk-llama when WHISPER_CUBLAS=1 (#1672)
983e4bd
unverified

bobqianic commited on

examples : Revert CMakeLists.txt for talk-llama (#1669)
92a92ed
unverified

bobqianic commited on

cmake : set default CUDA architectures (#1667)
0969db5
unverified

bobqianic commited on

bench.py : add different large models (#1655)
282c3a3
unverified

Alfredo Montesinos commited on

wchess : update README.md
8401568
unverified

ggerganov commited on

release : v1.5.2
9dd774b
unverified

ggerganov commited on

wchess : update readme
4b1cef7
unverified

ggerganov commited on

wchess : whisper assisted chess (#1595)
6260b52
unverified

fraxy-v ggerganov commited on

sync : ggml (Metal fixes, new ops, tests) (#1633)
a0d4b48
unverified

ggerganov commited on

cmake : target windows 8 or above for prefetchVirtualMemory in llama-talk (#1617)
ed75714
unverified

Kreijstal commited on

cmake : Fix bug in httplib.h for mingw (#1615)
334a75d
unverified

Kreijstal commited on

metal : fix `ggml_metal_log` vargs (#1606)
b3cea90
unverified

finnvoorhees commited on

whisper.objc : disable timestamps for real-time transcription
fc6989a
unverified

ggerganov commited on

whisper : more debug messages + fix fallback logic
51d2d34
unverified

ggerganov commited on

metal : fix soft_max kernel src1 argument (#1602)
5692844
unverified

ggerganov commited on

sync : ggml (new ops, new backend, etc) (#1602)
895e87a
unverified

ggerganov commited on

server : pass max-len argument to the server (#1574)
5704da9
unverified

osdrv commited on

ios : Remove `#if arch(arm)` check for using Metal (#1561)
3a600be
unverified

finnvoorhees commited on

ggml : Fix 32-bit compiler warning (#1575)
3931068
unverified

Digipom commited on

ggml : re-enable blas for src0 != F32 (#1583)
87987de
unverified

ggerganov commited on

Server : Add support for .vtt format to Whisper server (#1578)
17f1f89
unverified

Aleksander Andrzejewski commited on

server : backport .srt output format (#1565)
77aa181
unverified

Oleg Sidorov commited on

cmake : install required ggml.h header (#1568)
9001318
unverified

Gregor Jasny commited on

server : set default CORS headers to allow all (#1567)
531e483
unverified

Kasumi commited on

readme : update help (#1560)
a5ad309
unverified

bebound commited on

CI : Add CUDA 11.8.0 support (#1554)
8eaa345
unverified

bobqianic commited on

CI : Rectify the Clang-Related workflow issues (#1551)
c093ec4
unverified

bobqianic commited on