 Georgi Gerganov
|
39173bcacb
context : reserve new scheduler when graph topology changes (#18547)
|
преди 2 седмици |
 Johannes Gäßler
|
64848deb18
llama-fit-params: free memory target per device (#18679)
|
преди 3 седмици |
 Julius Tischbein
|
2038101bd9
llama : add `use_direct_io` flag for model loading (#18166)
|
преди 3 седмици |
 Daniel Bevenius
|
d3dce4e0a5
sampling : add support for backend sampling (#17004)
|
преди 3 седмици |
 Xuan-Son Nguyen
|
cd78e57c3a
lora: count lora nodes in graph_max_nodes (#18469)
|
преди 1 месец |
 o7si
|
daa242dfc8
common: fix return value check for setpriority (#18412)
|
преди 1 месец |
|
Johannes Gäßler
|
026d2ad472
llama: fix magic number of 999 for GPU layers (#18266)
|
преди 1 месец |
|
Johannes Gäßler
|
147a521636
tool/ex/tests: consistently free ctx, then model (#18168)
|
преди 1 месец |
|
Johannes Gäßler
|
a2c199e479
common: clarify instructions for bug reports (#18134)
|
преди 1 месец |
|
Johannes Gäßler
|
b1f3a6e5db
llama: automatically set parameters not set by the user in such a way that maximizes GPU utilization (#16653)
|
преди 1 месец |
|
Georgi Gerganov
|
254098a279
common : refactor common_sampler + grammar logic changes (#17937)
|
преди 1 месец |
 Sigbjørn Skjæret
|
22577583a3
common : change --color to accept on/off/auto, default to auto (#17827)
|
преди 1 месец |
 Adrien Gallouët
|
83c1171529
common: use native MultiByteToWideChar (#17738)
|
преди 1 месец |
 Reese Levine
|
7ca5991d2b
ggml webgpu: add support for emscripten builds (#17184)
|
преди 1 месец |
|
Xuan-Son Nguyen
|
13628d8bdb
server: add --media-path for local media files (#17697)
|
преди 1 месец |
|
Xuan-Son Nguyen
|
ec18edfcba
server: introduce API for serving / loading / unloading multiple models (#17470)
|
преди 1 месец |
 Aaron Teo
|
877566d512
llama: introduce support for model-embedded sampling parameters (#17120)
|
преди 2 месеца |
|
Georgi Gerganov
|
196f5083ef
common : more accurate sampling timing (#17382)
|
преди 2 месеца |
|
Xuan-Son Nguyen
|
9b17d74ab7
mtmd: add mtmd_log_set (#17268)
|
преди 2 месеца |
|
Xuan-Son Nguyen
|
aa3b7a90b4
arg: add --cache-list argument to list cached models (#17073)
|
преди 2 месеца |
 Gadflyii
|
3df2244df4
llama : add --no-host to disable host buffers (#16310)
|
преди 3 месеца |
|
Aaron Teo
|
624207e676
devops: add s390x & ppc64le CI (#15925)
|
преди 4 месеца |
 Douglas Hanley
|
b5bd037832
llama : add support for qwen3 reranker (#15824)
|
преди 4 месеца |
 Uilian Ries
|
152729f884
common : add missing chrono header for common.cpp (#16211)
|
преди 4 месеца |
|
Johannes Gäßler
|
e81b8e4b7f
llama: use FA + max. GPU layers by default (#15434)
|
преди 5 месеца |
|
Sigbjørn Skjæret
|
84ab83cc0b
model : jina-embeddings-v3 support (#13693)
|
преди 5 месеца |
|
Georgi Gerganov
|
9ebebef62f
llama : remove KV cache defragmentation logic (#15473)
|
преди 5 месеца |
 Jie Fu (傅杰)
|
2f3dbffb17
common : fix incorrect print of non-ascii characters in the logging (#15466)
|
преди 5 месеца |
 Jonathan Graehl
|
5cdb27e091
finetune: SGD optimizer, more CLI args (#13873)
|
преди 5 месеца |
 Diego Devesa
|
d6818d06a6
llama : allow other bufts when overriding to CPU, add --no-repack option (#14990)
|
преди 6 месеца |
