 0cc4m
|
e920ed393d
Vulkan Intel Fixes, Optimizations and Debugging Flags (#5301)
|
1 anno fa |
 Michael Klimenko
|
52bb63c708
refactor : switch to emplace_back to avoid extra object (#5291)
|
1 anno fa |
 Jared Van Bortel
|
1ec3332ade
YaRN : store rope scaling type as int32_t in memory (#5285)
|
1 anno fa |
 BADR
|
6a66c5071a
readme : add tenere in the ui tools list (#5284)
|
1 anno fa |
 AidanBeltonS
|
a305dba8ff
Fix im2col with 32fp (#5286)
|
1 anno fa |
 kalomaze
|
191221178f
perplexity : fix KL divergence calculations on Windows (#5273)
|
1 anno fa |
 Georgi Gerganov
|
e437b37fd0
scripts : parse wtype in server-llm.sh (#5167)
|
1 anno fa |
 Mirror Azure
|
2d40085c26
py : add check for '.attn.masked_bias' layers to GPT2model (#5281)
|
1 anno fa |
|
AidanBeltonS
|
b05102fe8c
Tidy ggml-sycl (#5261)
|
1 anno fa |
 Xuan Son Nguyen
|
6b91b1e0a9
docker : add build for SYCL, Vulkan + update readme (#5228)
|
1 anno fa |
 Meng, Hengyu
|
e805f0fa99
[SYCL] get MAX_MEM_ALLOC from device property (#5270)
|
1 anno fa |
 Neo Zhang Jianyu
|
af3ba5d946
[SYCL] update guide of SYCL backend (#5254)
|
1 anno fa |
 Ian Bull
|
e1e721094d
llama : fix memory leak in llama_batch_free (#5252)
|
1 anno fa |
|
Neo Zhang Jianyu
|
128dcbd3c9
add --no-mmap in llama-bench (#5257)
|
1 anno fa |
|
0cc4m
|
4d0924a890
Vulkan Phi Fix for AMD Proprietary Drivers (#5260)
|
1 anno fa |
 slaren
|
8ca511cade
cuda : fix LLAMA_CUDA_F16 (#5262)
|
1 anno fa |
 Ali Nehzat
|
d71ac90985
make : generate .a library for static linking (#5205)
|
1 anno fa |
 Guoteng
|
ce32060198
llama : support InternLM2 (#5184)
|
1 anno fa |
 Eve
|
1cfb5372cf
Fix broken Vulkan Cmake (properly) (#5230)
|
1 anno fa |
|
Georgi Gerganov
|
d3bac7d584
llama : reorder build_orion() at correct place (#5118)
|
1 anno fa |
|
Georgi Gerganov
|
5cb04dbc16
llama : remove LLAMA_MAX_DEVICES and LLAMA_SUPPORTS_GPU_OFFLOAD (#5240)
|
1 anno fa |
|
Georgi Gerganov
|
efb7bdbbd0
metal : add im2col F32 dst support (#5132)
|
1 anno fa |
 JidongZhang-THU
|
15606309a0
llava : add MobileVLM support (#5132)
|
1 anno fa |
|
Neo Zhang Jianyu
|
b2b9f025e7
format license text, restore apache license by legal suggestion (#5233)
|
1 anno fa |
|
slaren
|
dabcc5b471
ggml : limit n_threads to the max n_tasks (#5238)
|
1 anno fa |
|
0cc4m
|
f8e9140cb4
Vulkan Fixes (#5223)
|
1 anno fa |
 Yiming Cui
|
d62520eb2c
Fix typos of IQ2_XXS and IQ3_XXS in llama.cpp (#5231)
|
1 anno fa |
|
Neo Zhang Jianyu
|
01684139c3
support SYCL backend windows build (#5208)
|
1 anno fa |
|
Jared Van Bortel
|
e8dc55d006
kompute : llama-bench support and ggml_cpu_has_kompute() (#5226)
|
1 anno fa |
|
Georgi Gerganov
|
e0085fdf7c
Revert "server : change deps.sh xxd files to string literals (#5221)"
|
1 anno fa |
