 Sigbjørn Skjæret
|
3678b838bb
llama : support GEGLU for jina-bert-v2 (#14090)
|
7 달 전 |
 Jeff Bolz
|
652b70e667
vulkan: force device 0 in CI (#14106)
|
7 달 전 |
 Juk Armstrong
|
3a12db23b6
Fixed spec timings to: accepted/tested instead of accepted/drafted (#14104)
|
7 달 전 |
 Georgi Gerganov
|
ae92c1855b
sync : ggml
|
7 달 전 |
|
Georgi Gerganov
|
b7ce1ad1e3
ggml : fix weak alias win32 (whisper/0)
|
7 달 전 |
 0cc4m
|
97340b4c99
Vulkan: Don't default to CPU device (like llvmpipe), even if no other device is available, to allow fallback to CPU backend (#14099)
|
7 달 전 |
 Isaac McFadyen
|
2bb0467043
rpc : nicer error messages for RPC server crash (#14076)
|
7 달 전 |
|
Georgi Gerganov
|
b8e2194efc
sync : ggml
|
7 달 전 |
 Kai Pastor
|
1a3b5e80f7
Add in-build ggml::ggml ALIAS library (ggml/1260)
|
7 달 전 |
|
Georgi Gerganov
|
1f63e75f3b
metal : use less stack memory in FA kernel (#14088)
|
7 달 전 |
|
Georgi Gerganov
|
40cbf571c9
kv-cache : fix shift and defrag logic (#14081)
|
7 달 전 |
 Diego Devesa
|
7f4fbe5183
llama : allow building all tests on windows when not using shared libs (#13980)
|
7 달 전 |
 xctan
|
f470bc36be
ggml-cpu : split arch-specific implementations (#13892)
|
7 달 전 |
|
Diego Devesa
|
8f47e25f56
cuda : fix device sync on buffer clear (#14033)
|
7 달 전 |
|
Georgi Gerganov
|
201b31dc2e
graph : fix geglu (#14077)
|
7 달 전 |
 Xinpeng Dou
|
e21d2d4ae2
CANN: Simplify the environment variable setting(#13104)
|
7 달 전 |
 R0CKSTAR
|
dc0623fddb
webui: fix sidebar being covered by main content (#14082)
|
7 달 전 |
|
Georgi Gerganov
|
87d34b381d
server : fix LRU check (#14079)
|
7 달 전 |
 Nicolò Scipione
|
b460d16ae8
sycl: Add reorder to Q6_K mmvq implementation (#13885)
|
7 달 전 |
 Đinh Trọng Huy
|
91a8ee6a6f
add geglu activation function (#14074)
|
7 달 전 |
 Yuanhao Ji
|
056eb74534
CANN: Enable labeler for Ascend NPU (#13914)
|
7 달 전 |
|
Diego Devesa
|
247e5c6e44
cuda : fix buffer type check with integrated GPUs (#14069)
|
7 달 전 |
 吴小白
|
5787b5da57
ci: add LoongArch cross-compile build (#13944)
|
7 달 전 |
 Akarshan Biswas
|
228f34c9ce
SYCL: Implement few same quantized type copy kernels (#13739)
|
7 달 전 |
|
Sigbjørn Skjæret
|
0974ad7a7c
llama : fix llama_model_chat_template with template name (LLM_KV with suffix) (#14050)
|
7 달 전 |
|
Georgi Gerganov
|
745aa5319b
llama : deprecate llama_kv_self_ API (#14030)
|
7 달 전 |
|
Georgi Gerganov
|
487a5e0401
context : fix SWA-related warning for multiple sequences (#14045)
|
7 달 전 |
|
Sigbjørn Skjæret
|
d17a809ef0
llama : support multiple classifier outputs and labels (#13940)
|
7 달 전 |
|
Sigbjørn Skjæret
|
1caae7fc6c
gguf-py : add add_classifier_output_labels method to writer (#14031)
|
7 달 전 |
 Masato Nakasaka
|
669c13e0f6
vulkan: Enable VK_KHR_cooperative_matrix extension for Intel Xe2 GPUs (#14001)
|
7 달 전 |
