 Jeff Bolz
|
bf2a99e3cb
vulkan: update windows SDK in release.yml (#14344)
|
7 months ago |
 Molly Sophia
|
72c6bc3f3d
llama : better rwkv chat template and add missing `inputs.use_jinja` setting (#14336)
|
7 months ago |
 Johannes Gäßler
|
defe2158dd
CUDA: mul_mat_v support for batch sizes > 1 (#14262)
|
7 months ago |
 Georgi Gerganov
|
7b50d589a8
kv-cells : fix tracking of seq_pos (#14339)
|
7 months ago |
|
Jeff Bolz
|
3a9457df96
vulkan: update windows SDK in CI (#14334)
|
7 months ago |
 Ed Addario
|
fa4a9f2a1c
quantize : handle user-defined pruning of whole layers (blocks) (#13037)
|
7 months ago |
 Sigbjørn Skjæret
|
238005c2dc
gguf-py : fix SpecialVocab parsing when post_processor is null (#14330)
|
7 months ago |
 Ruikai Peng
|
66aba7aca9
run : avoid double tokenization (#14327)
|
7 months ago |
|
Georgi Gerganov
|
f1f5e82df6
examples : fix is_first logic for tokenization (#14329)
|
7 months ago |
 uvos
|
af3373f1ad
HIP: enable vec fattn on RDNA4 (#14323)
|
7 months ago |
 yuiseki
|
5d5c066de8
mtmd : fix Pixtral OOM with large images by capping image_size to 1024 (#14326)
|
7 months ago |
|
Sigbjørn Skjæret
|
40bfa04c95
common : use std::string_view now that we target c++17 (#14319)
|
7 months ago |
 Aman Gupta
|
aa064b2eb7
CUDA: add mean operation (#14313)
|
7 months ago |
|
Sigbjørn Skjæret
|
aa0ef5c578
gguf-py : fix Qwen3-Embedding eos token (#14314)
|
7 months ago |
 Markus Tavenrath
|
bb16041cae
Add support for VK_EXT_debug_utils to add labels to Vulkan objects. (#13792)
|
7 months ago |
|
Sigbjørn Skjæret
|
58cba76a9a
gguf-py : fix TemplateProcessing pair when bos/eos is missing (#14312)
|
7 months ago |
|
Georgi Gerganov
|
67ae5312e2
metal : fix thread-safety (#14300)
|
7 months ago |
|
Georgi Gerganov
|
692e3cdd0a
memory : rename interface to llama_memory_context_i (#14296)
|
7 months ago |
 Daniel Han
|
b23fa0b3f4
convert : fix Llama 4 conversion (#14311)
|
7 months ago |
|
Georgi Gerganov
|
06cbedfca1
sync : ggml
|
7 months ago |
 Acly
|
b7147673f2
Add `ggml_roll` (ggml/1274)
|
7 months ago |
 David Chiu
|
d860dd99a4
docs : fix the link to llama.h (#14293)
|
7 months ago |
|
Aman Gupta
|
c959f462a0
CUDA: add conv_2d_transpose (#14287)
|
7 months ago |
|
Sigbjørn Skjæret
|
22015b2092
lint : remove trailing whitepace (#14304)
|
7 months ago |
|
Ruikai Peng
|
dd6e6d0b6a
vocab : prevent tokenizer overflow (#14301)
|
7 months ago |
 Nicolò Scipione
|
8308f98c7f
sycl: add usage of enqueue_functions extension (#14244)
|
7 months ago |
 Christian Kastner
|
6369be0735
Implement GGML_CPU_ALL_VARIANTS for PowerPC (#14286)
|
7 months ago |
|
Sigbjørn Skjæret
|
88fc854b4b
llama : improve sep token handling (#14272)
|
7 months ago |
 Diego Devesa
|
e28c1b93fd
cuda : synchronize graph capture and cublas handle destruction (#14288)
|
7 months ago |
|
Georgi Gerganov
|
d27b3ca175
ggml : fix repack work size for mul_mat_id (#14292)
|
7 months ago |
