aa956
|
d67341dc18
server : add server parameters for draft model cache type (#13782)
|
преди 7 месеца |
fanyang
|
456af35eb7
build : suppress gcc15 compile warnings (#14261)
|
преди 7 месеца |
Anton Mitkov
|
600e3e9b50
sycl: Cleanup codepaths in Get Rows in sycl backend (#14215)
|
преди 7 месеца |
bashayer hijji
|
fffcce535e
llama-bench : add --no-warmup flag (#14224) (#14270)
|
преди 7 месеца |
pqnet
|
5fc7856815
convert : fix remote option in Windows (#14100)
|
преди 7 месеца |
Aaron Teo
|
faed5a5f5d
llamafile : support s390x SIMD instruction set (#14273)
|
преди 7 месеца |
0cc4m
|
10bb545c5b
Vulkan: Set device max size for host memory to avoid OOM warning and fallback to CPU buffer (#14249)
|
преди 7 месеца |
Gabe Goodhart
|
edc4a29eff
memory : Hybrid recurrent cache (#13979)
|
преди 7 месеца |
Georgi Gerganov
|
ed3290ab34
metal : add mean kernel (#14267)
|
преди 7 месеца |
Aaron Teo
|
8d94713654
docs: add s390x build documentation (#14264)
|
преди 7 месеца |
Aaron Teo
|
50d2227953
ggml-cpu: reduce asm calls for hsum (#14037)
|
преди 7 месеца |
Aaron Teo
|
6231c5cd6d
ggml-cpu: fix uncaught underscore terminators (#14023)
|
преди 7 месеца |
Charles Xu
|
ef035803eb
ggml: Add Apple support for GGML_CPU_ALL_VARIANTS (#14258)
|
преди 7 месеца |
Xuan-Son Nguyen
|
413977de32
mtmd : refactor llava-uhd preprocessing logic (#14247)
|
преди 7 месеца |
Xuan-Son Nguyen
|
95402553a5
llama-chat : fix multiple system message for gemma, orion (#14246)
|
преди 7 месеца |
Sigbjørn Skjæret
|
3865cff4f5
convert : fix null head_dim AutoConfig regression (#14248)
|
преди 7 месеца |
Georgi Gerganov
|
d03172cc79
sync : ggml
|
преди 7 месеца |
Daniel Bevenius
|
dd8e59f443
ggml : disable warnings for tests when using MSVC (ggml/1273)
|
преди 7 месеца |
Daniel Bevenius
|
bbe98d2784
ggml : remove unused ggml_context_container (ggml/1272)
|
преди 7 месеца |
Daniel Bevenius
|
c2056ed6d4
examples : include examples in msvc disable warn (ggml/1270)
|
преди 7 месеца |
bandoti
|
c46503014d
cmake: remove shader-gen step-targets from ggml-vulkan (#14226)
|
преди 7 месеца |
xctan
|
860a9e4eef
ggml-cpu : remove the weak alias trick (#14221)
|
преди 7 месеца |
R0CKSTAR
|
fe9d60e74a
musa: fix build warning (unused variable) (#14231)
|
преди 7 месеца |
Sigbjørn Skjæret
|
e434e69183
common : suggest --jinja when autodetection fails (#14222)
|
преди 7 месеца |
Georgi Gerganov
|
89fea80d29
server : fix incorrect usage of llama_get_embeddings() (#14225)
|
преди 7 месеца |
Diego Devesa
|
6adc3c3ebc
llama : add thread safety test (#14035)
|
преди 7 месеца |
bandoti
|
0dbcabde8c
cmake: clean up external project logic for vulkan-shaders-gen (#14179)
|
преди 7 месеца |
Đinh Trọng Huy
|
ad590be98c
model : add NeoBERT (#14164)
|
преди 7 месеца |
uvos
|
7d6d91babf
HIP: disable rocwmma on gfx12 by default until rocm 7.0 (#14202)
|
преди 7 месеца |
Georgi Gerganov
|
d3e64b9f49
llama : rework embeddings logic (#14208)
|
преди 7 месеца |