cmake : make libwhisper.so position independent (#1792)
1cf1553
unverified
trixirt
commited on
Spaces:
Running
Running
Commit History
cmake : temporary remove VLA check (#1795)
1a32e6f
unverified
whisper.android : return output from benchmarks (#1785)
5cff61b
unverified
server : implement "verbose_json" format with token details (#1781)
d6e13b6
unverified
ggml : sync ggml-metal.m
b4085c3
unverified
sync : llama.cpp
5de718a
unverified
sync : ggml
34bdd70
unverified
ggml : add IQ2 to test-backend-ops + refactoring (llama/4990)
227f2ae
unverified
imatrix : offload to GPU support (llama/4957)
6490f98
unverified
backend : add eval callback (llama/4935)
3cc64d6
unverified
metal : create autorelease pool during library build (llama/4970)
9027276
unverified
ggml : importance matrix support for legacy quants (llama/4969)
d8bb9d8
unverified
metal : log `recommendedMaxWorkingSetSize` on iOS 16+ (llama/4936)
e2cc0e5
unverified
ggml : introduce GGML_CALL function annotation (llama/4850)
7815f68
unverified
cuda : fix dequantize kernel names (llama/4938)
95f6502
unverified
CUDA: faster dequantize kernels for Q4_0 and Q4_1 (llama/4938)
73c6598
unverified
Add ability to use importance matrix for all k-quants (llama/4930)
7032309
unverified
talk-llama : optional wake-up command and audio confirmation (#1765)
542e8da
unverified
server : fix building and simplify lib deps on Windows (#1772)
f928f33
unverified
Przemysław Pawełczyk
commited on
talk-llama : sync llama.cpp
62ad8e0
unverified
talk-llama : llama.cpp
d128cb3
unverified
sync : ggml
6a472b5
unverified
metal : correctly set SIMD support flags on iOS (llama/4923)
1cf2fa9
unverified
2-bit quantizations (llama/4897)
8a399ab
unverified
scripts : sync-ggml-am.sh add option to skip commits
c34dd82
unverified
talk-llama : sync llama.cpp
b9d2bd9
unverified
sync : ggml
18bfc83
unverified
examples : adapt to metal API
b65decb
unverified
ggml: cache sin/cos for RoPE (llama/4908)
c315fbf
unverified
metal : remove old API (llama/4919)
d6abb6a
unverified
metal : disable log for loaded kernels (llama/4794)
2305485
unverified
gguf : fix potential infinite for-loop (llama/4600)
0e93179
unverified
texmex76
Bernhard Gstrein
commited on
metal : refactor kernel loading code (llama/4794)
53e6bf8
unverified
CUDA: faster q8_0 -> f16 dequantization (llama/4895)
0a1a178
unverified
talk-llama : add optional CLI arg to set the bot name (#1764)
63c8089
unverified
RhinoDevel
commited on
examples : add python example for transcription (#1744)
d600e4c
unverified
whisper : load the model into multiple buffers of max size 1GB (#1763)
0e9101f
unverified
talk-llama : sync llama.cpp
75c5f9c
unverified
sync : ggml
2ed0a44
unverified
backend_sched : fix assignments
cb91db5
unverified
slaren
commited on
llama : ggml-backend integration (llama/4766)
362430b
unverified
CUDA: fix softmax compile for old CUDA versions (llama/4862)
5eda533
unverified
models : make all scripts to be POSIX Compliant (#1725)
f7aef3e
unverified
ggml : fix 32-bit ARM compat for IQ2_XS (#1758)
d5836c9
unverified
go : add SetInitialPrompt method to bindings (#1753)
5fd6678
unverified
server : add more parameters to server api (#1754)
cb0cf7b
unverified
George Hindle
commited on
whisper : fix segment length with params.no_timestamps == true
720d738
unverified
params : don't compute timestamps when not printing them (#1755)
251825e
unverified
George Hindle
commited on