Commits

History for

Commits on Apr 8, 2024

remove row=1 cond (ggerganov#6532 )
abhilash1910
committedApr 8, 2024

Commits on Apr 7, 2024

support/fix OPs GGML_TYPE_IQ4_NL, GGML_TYPE_IQ4_XS, GGML_TYPE_IQ3_XXS, GGML_TYPE_IQ3_S, GGML_TYPE_IQ2_XXS, GGML_TYPE_IQ2_XS, GGML_TYPE_IQ2_S, GGML_TYPE_IQ1_S, GGML_TYPE_IQ1_M (ggerganov#6521 )
NeoZhangJianyu
committedApr 7, 2024

Commits on Apr 5, 2024

[SYCL] Fixed minor bug when enabling FP16 for non intel targets (ggerganov#6464 )

OuadiElfarouki
and
AidanBeltonS
committedApr 5, 2024

Commits on Apr 3, 2024

[SYCL] Disable iqx on windows as WA (ggerganov#6435 )
airMeng
committedApr 3, 2024

Commits on Mar 28, 2024

[SYCL] fix set main gpu crash (ggerganov#6339 )
NeoZhangJianyu
committedMar 28, 2024

Commits on Mar 27, 2024

[SYCL] Fix batched impl for NVidia GPU (ggerganov#6164 )

AidanBeltonS
and
abhilash1910
committedMar 27, 2024

Commits on Mar 26, 2024

llama : greatly reduce output buffer memory usage (ggerganov#6122 )

compilade
and
ggerganov
committedMar 26, 2024

Commits on Mar 24, 2024

[SYCL] offload op (ggerganov#6217 )
airMeng
committedMar 24, 2024

Commits on Mar 21, 2024

Add nvidia and amd backends (ggerganov#6157 )
AidanBeltonS
committedMar 21, 2024

Commits on Mar 18, 2024

backend : offload large batches to GPU (ggerganov#6083 )

slaren
and
JohannesGaessler
committedMar 18, 2024

Commits on Mar 15, 2024

fix set main gpu error (ggerganov#6073 )
NeoZhangJianyu
committedMar 15, 2024
[SYCL] Fix non-intel device selection (ggerganov#6042 )

committedMar 15, 2024

Commits on Mar 13, 2024

llama : add pipeline parallelism support (ggerganov#6017 )

committedMar 13, 2024
Update get version (ggerganov#6025 )
AidanBeltonS
committedMar 13, 2024

Commits on Mar 12, 2024

ggml : reuse quantum structs across backends (ggerganov#5943 )
ggerganov
committedMar 12, 2024
sycl : update IQ1_S kernels (WIP - not working!) (ggerganov#5995 )
ggerganov
committedMar 12, 2024

Commits on Mar 11, 2024

[SYCL] Add q3_s and q1_s (ggerganov#5886 )
abhilash1910
committedMar 11, 2024

Commits on Mar 9, 2024

ggml : add ggml-common.h to deduplicate shared code (ggerganov#5940 )
ggerganov
committedMar 9, 2024

Commits on Mar 7, 2024

Commits on Mar 6, 2024

add wait() to make code stable (ggerganov#5895 )
NeoZhangJianyu
committedMar 6, 2024

Commits on Mar 5, 2024

[SYCL] fix mul_mat fault in CI/unit-test (ggerganov#5862 )

committedMar 5, 2024

Commits on Mar 4, 2024

ggml : introduce ggml_status (ggml/750)

committedMar 4, 2024

Commits on Mar 2, 2024

Support multiple GPUs (split mode) on SYCL backend (ggerganov#5806 )
NeoZhangJianyu
committedMar 2, 2024

Commits on Mar 1, 2024

[SYCL] Use batched mul_mat pathway (ggerganov#5591 )

AidanBeltonS
and
abhilash1910
committedMar 1, 2024

Commits on Feb 28, 2024

Introduce backend GUIDs (ggml/743)

authored and
ggerganov
committedFeb 28, 2024

Commits on Feb 26, 2024

[SYCL] Add support for soft_max ALiBi (ggerganov#5639 )

AidanBeltonS
and
abhilash1910
committedFeb 26, 2024

Commits on Feb 25, 2024

code : normalize enum names (ggerganov#5697 )
ggerganov
committedFeb 25, 2024

Commits on Feb 21, 2024

[SYCL] conext add name (ggerganov#5624 )
airMeng
committedFeb 21, 2024

Commits on Feb 20, 2024

Update ggml_sycl_op_mul_mat_vec_q (ggerganov#5502 )

AidanBeltonS
and
abhilash1910
committedFeb 20, 2024

Commits on Feb 12, 2024

ggml-sycl: Replace 3d ops with macro (ggerganov#5458 )
abhilash1910
committedFeb 12, 2024

Commits on Feb 8, 2024

Fix f16_sycl cpy call from Arc (ggerganov#5411 )
abhilash1910
committedFeb 8, 2024

Commits on Feb 5, 2024

[SYCL] Fix cpy with dims of 3 (ggerganov#5289 )

AidanBeltonS
and
abhilash1910
committedFeb 5, 2024

Commits on Feb 3, 2024

Fix im2col with 32fp (ggerganov#5286 )
AidanBeltonS
committedFeb 3, 2024

Commits on Feb 2, 2024

Tidy ggml-sycl (ggerganov#5261 )

AidanBeltonS
and
abhilash1910
committedFeb 2, 2024