Spaces:
Running
Running
Commit History
server : add option to suppress non-speech tokens (#2649)
647c7e7
unverified
whisper : rename binaries + fix install (#2648)
30197de
unverified
ruby : update gem version to v1.3.1
95fda6c
unverified
release : v1.7.3
50c0d82
unverified
ci : msys enable SDL2 build (#2635)
cf6eb54
unverified
ruby : sync ggml (#2643)
916c6e0
android : try to fix build
c9db590
files : remove old sources
1da9474
sync : ggml
9442640
talk-llama : sync llama.cpp
c4fb34c
sync : ggml
3d08664
ggml : update ggml_backend_cpu_device_supports_op (llama/10867)
2f11d1e
vulkan: bugfixes for small subgroup size systems + llvmpipe test (llama/10809)
9220b51
Eve
commited on
rwkv6: add wkv6 support for Vulkan backend (llama/10829)
c7285d6
Introducing experimental OpenCL backend with support for Qualcomm Adreno GPUs (llama/10693)
83a0899
lhez
Skyler Szot
Shangqing Gu
Alexander Angus
Hongqiang Wang
Max Krasnyansky
commited on
Fix crash caused by ggml_backend_load_all when launching on Android Activity (llama/10812)
e1df33d
谢乃闻
Diego Devesa
commited on
vulkan: small mul_mat_vec optimizations (llama/10665)
ec98109
Eve
commited on
SYCL: Reduce most of the compiler warnings (llama/10748)
050e6ce
ggml : Fix compilation issues on ARM platform when building without fp16 (llama/10811)
f76ba41
Karol Kontny
commited on
CUDA: faster non-contiguous concat (llama/10760)
4621719
a3sh
Diego Devesa
commited on
remove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS (llama/10797)
b38cecf
Diego Devesa
commited on
Vulkan: Use improved q4_k and q5_k dequant code in dequant shaders (llama/10798)
a812efc
Vulkan: Add VK_EXT_subgroup_size_control support to ensure full subgroups for coopmats (llama/10721)
488f19e
ggml: load all backends from a user-provided search path (llama/10699)
c6de218
Gilad S
Diego Devesa
commited on
vulkan: request round-to-even for fp16 in im2col/rope_head (llama/10767)
461484c
vulkan: dynamic subgroup size for the remaining k quants (llama/10745)
1bbdb81
Eve
commited on
CUDA: rename macros to avoid conflicts with WinAPI (llama/10736)
8544072
Andreas Kieslinger
commited on
vulkan: disable spirv-opt for coopmat shaders (llama/10763)
2ac53b2
ggml : remove return from ggml_gallocr_allocate_node (ggml/1048)
f9d4408
ggml : add check for grad_accs (ggml/1046)
eacc95c
common : remove old types
fc4a926
CUDA: fix shared memory access condition for mmv (llama/10740)
99a4546
vulkan: fix compile warnings (llama/10731)
cdcb67c
Vulkan: fix NaN in tanh.comp with AMD proprietary driver on Windows (llama/10723)
a618c84
stduhpf
commited on
vulkan: compile a test shader in cmake to check for coopmat2 support (llama/10713)
980eeb3
ggml : disable iq4_nl interleave size 8 (llama/10709)
a5294e7
ggml : refactor online repacking (llama/10446)
163128e
Vulkan: VK_KHR_cooperative_matrix support to speed up prompt processing (llama/10597)
9a4de04
metal : Extend how Llama.cpp locates metal resources (llama/10676)
44e7250
vulkan: Add VK_NV_cooperative_matrix2 support for mul_mat and flash attention (llama/10206)
d10b47b
ruby : Add no_speech_thold (#2641)
91607b6
unverified
stream : improve consistency in README (#2642)
91a639e
unverified
crummyh
commited on
whisper : support no_speech_thold (#2625)
adb5837
unverified
Karthick
commited on
whisper : add single-timestamp logic (#2629)
7655c06
unverified
readme : fix typo (#2637)
7fd5b82
unverified
crummyh
commited on
