 Johannes Gäßler
|
8e6e2fbe14
CUDA: fix variable name conflict for Windows build (#9382)
|
преди 1 година |
 Antonis Makropoulos
|
5ed087573e
readme : add LLMUnity to UI projects (#9381)
|
преди 1 година |
 Radoslav Gerganov
|
54f376d0b9
rpc : update README [no ci] (#9320)
|
преди 1 година |
 Dan Johansson
|
b2e89a3274
Arm AArch64: Documentation updates (#9321)
|
преди 1 година |
 Markus Tavenrath
|
daa9623ab0
Overlap cmdbuffer creation and cmdbuffer execution in Vulkan backend by submitting smaller cmdbuffers early. (#9118)
|
преди 1 година |
 Georgi Gerganov
|
e079bffb66
cuda : fix FA Q src index (1 -> 0) (#9374)
|
преди 1 година |
 Xuan Son Nguyen
|
3f7ccfd649
common : bring back missing args, add env var duplication check (#9375)
|
преди 1 година |
 slaren
|
a249843d89
common : restore --n-gpu-layers (#9371)
|
преди 1 година |
|
slaren
|
19f4a7b296
llama : refactor samplers internal implementation (#9370)
|
преди 1 година |
 Neo Zhang Jianyu
|
2a358fb0c4
[SYCL] add check malloc result on device (#9346)
|
преди 1 година |
|
slaren
|
eae597182c
llama : sanitize tokens in the upper bound (#9359)
|
преди 1 година |
|
Xuan Son Nguyen
|
00b02bb249
imatrix : fix arg parser for imatrix (#9366)
|
преди 1 година |
|
Georgi Gerganov
|
a876861455
metal : update support condition for im2col + fix warning (#0)
|
преди 1 година |
|
Georgi Gerganov
|
385decbd63
sync : ggml
|
преди 1 година |
|
Georgi Gerganov
|
60a3107ccd
scripts : option to increase git patch context
|
преди 1 година |
 Salvatore Mesoraca
|
406c1a32a1
vulkan: add dryrun support to sin and cos ops (ggml/947)
|
преди 1 година |
|
Salvatore Mesoraca
|
9cb9260861
vulkan: correctly report support for OP_CONT (ggml/946)
|
преди 1 година |
|
Johannes Gäßler
|
202084d31d
tests: add gradient tests for all backends (ggml/932)
|
преди 1 година |
|
Johannes Gäßler
|
dbbebcab33
ggml: fix ggml_graph_cpy undefined behavior (ggml/943)
|
преди 1 година |
|
Georgi Gerganov
|
ba1cf846ed
cann : fix doxy (ggml/0)
|
преди 1 година |
 Mengqing Cao
|
d2d3200b38
cann : add Ascend NPU support (whisper/2336)
|
преди 1 година |
|
Georgi Gerganov
|
51d964a4ef
cuda : mark BF16 CONT as unsupported
|
преди 1 година |
|
Salvatore Mesoraca
|
efe6a83e30
ggml : fix cont with transposed tensors when one dimension is 1 (ggml/934)
|
преди 1 година |
 Kevin Gibbons
|
fbb7fcffbc
llama : set attrs of mislabelled EOT/EOM tokens (#9348)
|
преди 1 година |
|
Georgi Gerganov
|
a5b5d9a101
llama.android : fix build (#9350)
|
преди 1 година |
|
Georgi Gerganov
|
f12295b8a9
llama : fix empty ring buffer push (#9358)
|
преди 1 година |
|
Georgi Gerganov
|
faf69d4237
llama : sanitize invalid tokens (#9357)
|
преди 1 година |
 Eve
|
e536426ded
llamafile : disable sgemm for batch-size 1 (#9330)
|
преди 1 година |
|
Xuan Son Nguyen
|
1b9ae5189c
common : refactor arg parser (#9308)
|
преди 1 година |
|
slaren
|
e32d0816ed
ggml : always check bounds on get_rows operations (#9354)
|
преди 1 година |
