Diego Devesa
|
b617cfd289
ggml-alloc : fix leak when reusing a tensor with a larger size (#16679)
|
2 месяцев назад |
Aleksander Grygier
|
79068501fa
Prevent premature submission on IME input (#16673)
|
2 месяцев назад |
Aleksander Grygier
|
0e4a0cf2fa
Import/Export UX improvements (#16619)
|
2 месяцев назад |
Aleksander Grygier
|
13f2cfad41
Enable per-conversation loading states to allow having parallel conversations (#16327)
|
2 месяцев назад |
takuya kodama
|
06332e2867
llama-batch: fix build fails with `-Werror=missing-braces` (#16614)
|
2 месяцев назад |
Ron Evans
|
72d53e6c4d
readme: update bindings (#16651)
|
2 месяцев назад |
safranowith
|
2330de7b84
SYCL: Add support for FLOOR,CEIL,ROUND and TRUNC unary operators (#16613)
|
2 месяцев назад |
takuya kodama
|
7062dd8460
llama-context: only warn on pooling_type when user specified (#16674)
|
2 месяцев назад |
Giuseppe Scrivano
|
0398752dd4
model : add Granite Hybrid types (#16635)
|
3 месяцев назад |
Aaron Teo
|
4f73d0a951
ci : fix binaries release failure for s390x (binaries may not work yet) (#16664)
|
3 месяцев назад |
Sigbjørn Skjæret
|
cec5edbcae
ci : avoid manual updates of docs/ops.md (#16663)
|
3 месяцев назад |
Aaron Teo
|
fcb235b466
ci: include s390x release binaries (#16648)
|
3 месяцев назад |
Aman Gupta
|
55754bebd5
CODEOWNERS: update for ggml-cuda/mmf (#16660)
|
3 месяцев назад |
Johannes Gäßler
|
ee09828cb0
HIP: fix GPU_TARGETS (#16642)
|
3 месяцев назад |
Jeff Bolz
|
e56abd2098
vulkan: Implement topk_moe fused shader, ported from CUDA (#16641)
|
3 месяцев назад |
Aman Gupta
|
38355c6c8e
CUDA: use registers instead of smem in topk-moe (#16647)
|
3 месяцев назад |
Shawn Gu
|
81387858f1
opencl: transposed gemm/gemv moe kernel with mxfp4,f32 (#16602)
|
3 месяцев назад |
Johannes Gäßler
|
66b0dbcb2d
llama-model: fix insonsistent ctxs <-> bufs order (#16581)
|
3 месяцев назад |
Radoslav Gerganov
|
41386cf365
rpc : report actual free memory (#16616)
|
3 месяцев назад |
Giuseppe Scrivano
|
3d4e86bbeb
vulkan: Add State Space Model (SSM) Operations Support (#16463)
|
3 месяцев назад |
muggle-stack
|
342c728d03
ggml : fix SpaceMit IME array out-of-bounds in task assignment (#16629)
|
3 месяцев назад |
Pascal
|
ababae7e1e
webui: reorganize settings layout (#16607)
|
3 месяцев назад |
Jeff Bolz
|
b19491599d
vulkan: fix debug build (add_rms_len/data not found) (#16624)
|
3 месяцев назад |
Ilia Ilmer
|
9ad4f1931e
metal : add `CONV_TRANSPOSE_2D` (#16542)
|
3 месяцев назад |
Olivier Chafik
|
79967ec596
grammar : use int64_t to avoid int overflows in int schema to grammar conversion logic (#16626)
|
3 месяцев назад |
GittyBurstein
|
ceff6bb253
SYCL SET operator optimized for F32 tensors (#16350)
|
3 месяцев назад |
Xuan-Son Nguyen
|
1bb4f43380
mtmd : support home-cooked Mistral Small Omni (#14928)
|
3 месяцев назад |
Pascal
|
683fa6ba4e
fix: added a normalization step for MathJax-style \[\] and \(\) delimiters (#16599)
|
3 месяцев назад |
GittyBurstein
|
b22572e97d
sycl : add ARANGE operator (#16362)
|
3 месяцев назад |
Chenguang Li
|
7a50cf388a
CANN: format code using .clang-format (#15863)
|
3 месяцев назад |