Georgi Gerganov
|
609a2d0268
models : fix YaRN regression + consolidate logic (#18006)
|
пре 1 месец |
Georgi Gerganov
|
a63cbafbbc
ggml : arm repack fix build
|
пре 1 месец |
Georgi Gerganov
|
0e59224990
sync : ggml
|
пре 1 месец |
Georgi Gerganov
|
71fdcf0616
ggml : arm repack fix build (whisper/0)
|
пре 1 месец |
Congcong Cai
|
615655aafe
cmake : set `CMAKE_RUNTIME_OUTPUT_DIRECTORY` for non standalone build (ggml/1394)
|
пре 1 месец |
Xuan-Son Nguyen
|
c00ff929dc
scripts: add script to compare logprobs of llama.cpp against other frameworks (#17947)
|
пре 1 месец |
Sergey Fedorov
|
4ed2bae50d
server-models.cpp: add missing <filesystem> (#18000)
|
пре 1 месец |
Jeff Bolz
|
5266379bca
llama_context: synchronize before reallocating output buffer (#17974)
|
пре 1 месец |
Xuan-Son Nguyen
|
4d5ae24c0a
arg: fix common_params_parse not accepting negated arg (#17991)
|
пре 1 месец |
Gustavo Rocha Dias
|
66ba51252e
cmake: correct scope - link ws2_32 for MinGW/w64devkit builds in cpp-httplib (#17972)
|
пре 1 месец |
Jeff Bolz
|
36255a2268
vulkan: support get_rows for i32 (#17941)
|
пре 1 месец |
Jeff Bolz
|
3229a23fa6
vulkan: support GGML_OP_DIAG (#17893)
|
пре 1 месец |
Jeff Bolz
|
303f8615e9
vulkan: Multi-pass softmax for large number of cols (#17892)
|
пре 1 месец |
Georgi Gerganov
|
3c6391e748
speculative-simple : free batch on exit (#17985)
|
пре 1 месец |
Sigbjørn Skjæret
|
8e4d678528
common : skip model validation when --completion-bash is requested (#17975)
|
пре 1 месец |
Jeff Bolz
|
07a10c1090
vulkan: Allow non-pow2 n_experts in topk_moe (#17872)
|
пре 1 месец |
Sigbjørn Skjæret
|
2bc94e7928
add llama-completion to completion-bash executables (#17976)
|
пре 1 месец |
Daniel Bevenius
|
fd1085ffb7
model-conversion : use CONVERTED_MODEL value for converted model [no ci] (#17984)
|
пре 1 месец |
Xuan-Son Nguyen
|
380b4c984e
common: support negated args (#17919)
|
пре 1 месец |
Xuan-Son Nguyen
|
e39a2ce66d
clip: move model cgraphs into their own files (#17965)
|
пре 1 месец |
jiahao su
|
a8c7f33d79
ci : change the cann version and the container pull method (#17953)
|
пре 1 месец |
Sigbjørn Skjæret
|
b7f5f46e03
docker : include legacy llama-completion binary (#17964)
|
пре 1 месец |
Johannes Gäßler
|
482211438d
CUDA: fix overflow in MMA kernel without stream-k (#17939)
|
пре 1 месец |
Georgi Gerganov
|
7bed317f53
models : fix the attn_factor for mistral3 graphs + improve consistency (#17945)
|
пре 1 месец |
Sigbjørn Skjæret
|
dcb7d17758
cann : fix ops broken by circular padding guard (#17825)
|
пре 1 месец |
ixgbe
|
51604435e8
ggml-cpu : fix RISC-V Q4_0 repack select and RVV feature reporting (#17951)
|
пре 1 месец |
Xuan-Son Nguyen
|
17158965ac
mtmd: explicitly forbidden inclusion of private header and libcommon (#17946)
|
пре 1 месец |
Aleksander Grygier
|
12280ae905
webui: Fix parsing non-LaTeX occurrencies of `\(` or `\)` (#17810)
|
пре 1 месец |
Xuan-Son Nguyen
|
54a0fee4b7
arg: add -mm and -mmu as short form of --mmproj and --mmproj-url (#17958)
|
пре 1 месец |
Daniel Bevenius
|
dada4c846d
model-conversion : remove max diff check in compare-logits [no ci] (#17954)
|
пре 1 месец |