Commits

History for

Commits on May 29, 2024

ggml : fix YARN + add tests + add asserts (ggerganov#7617 )
ggerganov
committedMay 29, 2024
cuda : non-cont concat support (ggerganov#7610 )
ggerganov
committedMay 29, 2024

Commits on May 28, 2024

ggml : generalize GGML_OP_CONCAT (ggerganov#7563 )
ggerganov
committedMay 28, 2024

Commits on May 27, 2024

update HIP_UMA ggerganov#7399 (ggerganov#7414 )

Djip007
and
slaren
committedMay 27, 2024

Commits on May 23, 2024

ggml : drop support for QK_K=64 (ggerganov#7473 )
ggerganov
committedMay 23, 2024

Commits on May 22, 2024

CUDA: fix FA out-of-bounds reads (ggerganov#7479 )
JohannesGaessler
committedMay 22, 2024
CUDA: fix FA out-of-bounds writes (ggerganov#7465 )
JohannesGaessler
committedMay 22, 2024
cuda : fix compile warning (ggerganov#7454 )
ggerganov
committedMay 22, 2024
CUDA: remove incorrect precision check (ggerganov#7454 )
JohannesGaessler
committedMay 22, 2024
cuda : fix rope + add tests (ggerganov#7452 )
ggerganov
committedMay 22, 2024

Commits on May 21, 2024

llama : add phi3 128K model support (ggerganov#7225 )

liuwei-git
and
ggerganov
committedMay 21, 2024
CUDA: fix unused warning in mmq.cu (ggerganov#7442 )
JohannesGaessler
committedMay 21, 2024
CUDA: deduplicate mmq code (ggerganov#7397 )
JohannesGaessler
committedMay 21, 2024

Commits on May 18, 2024

CUDA: deduplicate FlashAttention code (ggerganov#7352 )
JohannesGaessler
committedMay 18, 2024
cuda : add half2 __shfl_xor() for ROCm 5.5 (ggerganov#7263 )
Engininja2
committedMay 18, 2024

Commits on May 17, 2024

CUDA: faster large batch FA without tensor cores (ggerganov#7314 )
JohannesGaessler
committedMay 17, 2024

Commits on May 15, 2024

ggml : add ggml_upscale_ext (ggml/814)

balisujohn
and
ggerganov
committedMay 15, 2024

Commits on May 12, 2024

CUDA: add FP32 FlashAttention vector kernel (ggerganov#7188 )
JohannesGaessler
committedMay 12, 2024

Commits on May 11, 2024

feat: implemented sigmoid function (ggml/806)

justcho5
authored and
ggerganov
committedMay 11, 2024
ggml : full ALiBi support (ggerganov#7192 )
ggerganov
committedMay 11, 2024

Commits on May 9, 2024

CUDA: generalize FP16 fattn vec kernel (ggerganov#7061 )
JohannesGaessler
committedMay 9, 2024

Commits on May 8, 2024

Introduction of CUDA Graphs to LLama.cpp (ggerganov#6766 )

agray3
and
slaren
committedMay 8, 2024

Commits on May 1, 2024

CUDA: CUDART < 11.7 workaround for __hmax, __hmax2 (ggerganov#7019 )
JohannesGaessler
committedMay 1, 2024

Commits on Apr 30, 2024

ggml : add Flash Attention (ggerganov#5021 )

committedApr 30, 2024

Commits on Apr 28, 2024

Fix more int overflow during quant (PPL/CUDA). (ggerganov#6563 )
dranger003
committedApr 28, 2024

Commits on Apr 18, 2024

ggml : group all experts in a single ggml_mul_mat_id (ggerganov#6505 )

slaren
and
ggerganov
committedApr 18, 2024

Commits on Apr 9, 2024

llama : add Command R Plus support (ggerganov#6491 )

committedApr 9, 2024

Commits on Apr 3, 2024

ggml : mul_mat_id use the same tensor for all the experts (ggerganov#6387 )

slaren
and
ggerganov
committedApr 3, 2024

Commits on Mar 29, 2024

sync : ggml (ggerganov#6351 )

ggerganov
and
slaren
committedMar 29, 2024

Commits on Mar 26, 2024

IQ1_M: 1.75 bpw quantization (ggerganov#6302 )

ikawrakow
and
Kawrakow
committedMar 26, 2024

Commits on Mar 25, 2024

cuda : fix LLAMA_CUDA_F16 build (ggerganov#6298 )
slaren
committedMar 25, 2024
cuda : refactor into multiple files (ggerganov#6269 )
slaren
committedMar 25, 2024

Loading