ggerganov/llama.cpp

Pull Requests Commits
llama : add "virtual sequences"
ggerganov committed 159 days ago
Verified dfceb012
ggml : add TODOs for adding GGML_OP_SET_ROWS support in the backends
ggerganov committed 159 days ago
Verified 30b4d4e1
kv-cache : add comments
ggerganov committed 159 days ago
Verified 5495ea96
kv-cache : bounds-check when accessing slot_info indices
ggerganov committed 159 days ago
Verified f3da97e6
kv-cache : improve find_slot impl
ggerganov committed 159 days ago
Verified a70293bc
cont : remove redundant ifs
ggerganov committed 159 days ago
Verified 2ac5be3a
graph : separate k and v indices
ggerganov committed 159 days ago
Verified ac8f3474
kv-cache : use ggml_set_rows
ggerganov committed 159 days ago
Verified cd811b7a
CUDA: add softmax broadcast (#14475)
am17an committed 159 days ago
55a1c5a5
CUDA: broadcasting for FlashAttention mask (#14500)
JohannesGaessler committed 159 days ago
12a81af4
vulkan: support softmax/FA batch and broadcast (#14449)
jeffbolznv committed 159 days ago
8875523e
ggml : support bcast ggml_soft_max_ext, ggml_flash_attn_ext (#14435)
ggerganov committed 159 days ago
ec68e84c
opencl : fix possible buffer overflow in dump_tensor (#14490)
jeffzhou2000 committed 159 days ago
Verified 307e79d3
simple-chat : fix context-exceeded condition (#14494)
ggerganov committed 159 days ago
Verified d7f5f4e5
opencl : skip empty nodes on cgraph compute (#14491)
EZForever committed 159 days ago
Verified c8a4e470
opencl : update upscale to support align corners (#14488)
lhez committed 159 days ago
Verified 603e43dc
ci : add OpenCL to labeler workflow (#14496)
CISC committed 159 days ago
Verified 611ba4b2
github : add OpenCL backend to issue templates (#14492)
EZForever committed 159 days ago
Verified 85841e12
ggml : Callback before abort (#14481)
ScaledLizard committed 159 days ago
Verified 68b3cd65
ci : disable fast-math for Metal GHA CI (#14478)
ggerganov committed 160 days ago
Verified de569441
Add Vulkan images to docker.md (#14472)
xek committed 160 days ago
Verified 1b2aaf28
CANN: update aclnnGroupedMatmulV2 to aclnnGroupedMatmulV3 (#14411)
noemotiovon committed 160 days ago
Verified 343b6e94
vulkan: Split large mul_mat_id to fit in shared memory (#14451)
jeffbolznv committed 160 days ago
Verified 6a746cf9
add GELU_ERF (#14455)
CISC committed 160 days ago
Verified eff5e454
ggml : remove trailing whitespace (#0)
ggerganov committed 160 days ago
a6a47958
sync : ggml
ggerganov committed 160 days ago
f61c05d4
ggml-cpu : "align corners" for bilinear upscale/downscale (ggml/1285)
Acly committed 160 days ago
431b2c24
ggml-quants : rename best_mad to best_error (ggml/1283)
danbev committed 160 days ago
497be7c0
opencl : add GEGLU, REGLU, SWIGLU (#14456)
lhez committed 160 days ago
Verified 79b33b23
Add Conv2d for CPU (#14388)
am17an committed 161 days ago
Verified 0a5a3b5c
Terms Of Use Privacy Policy Refund Policy Impressum