ggml-org/llama.cpp

Pull Requests Commits
ci : fix race in webgpu memset (#31162)
eapache committed 2 days ago
Verified 9125a2eb
metal : stop overflow from happening in large matrix output offsets (#30072)
feni6 committed 2 days ago
Verified 9b8446b9
ui: UX improvements (#31030)
allozaur committed 2 days ago
Verified dc9e3bed
ui: Sidebar resizing improvements (#31119)
allozaur committed 2 days ago
Verified 69b3cc40
ci : bump server timeout during startup (#31174)
ggerganov committed 2 days ago
Verified 224d94c0
ci : drop the `macos-latest-x64` job (#31175)
ggerganov committed 2 days ago
Verified 48c36110
deepseekv4.1: follow-up review fixes (#31137)
am17an committed 2 days ago
Verified a0742153
cuda : use std::string instead of VLA for archName (#30463)
angt committed 2 days ago
Verified d854d98b
common, grammar: add recursive depth limit, make it a requirement (#31153)
ngxson committed 2 days ago
Verified 48c9f08b
tests : adjust default threads for test-llama-context (#31157)
ggerganov committed 3 days ago
Verified 6050b467
qwen: merge tensors on convert (#31028)
am17an committed 3 days ago
Verified 2280f0fd
ci : disable test-llama-context with fa off in the webgpu (#30977)
yomaytk committed 3 days ago
Verified 1a0b56ce
llama: fix exhausting CPU and MEM caused by dry sequence breaker (#31035)
trunk-trick committed 3 days ago
Verified 63412d87
llama-quant : fix 4D tensor quantization using ggml_nrows() and name-based expert detection (#30820)
anthony-wss committed 3 days ago
Verified 5c965fd8
grammar: limit recursion depth to prevent stack overflow (#29495) (#29755)
Hrushikesh-ramilla committed 3 days ago
Verified f1168ac4
llama : respect no_alloc in all memory caches (#31087)
ggerganov committed 3 days ago
Verified fa3645fe
tools : reject truncated GGUF tensor data in merge (#31044)
shreya-11733 committed 3 days ago
Verified 7790a923
CUDA: fuse FFN gate/up matmuls and GLU in MMQ with interleaved gate/up rows (#29948)
anujj committed 3 days ago
Verified c49edf09
ggml-cuda: fuse qwen MRoPE + set-rows (#31025)
am17an committed 3 days ago
Verified a55c16e4
vulkan: Add a larger tile size for conv3d (coopmat2 only) (#29920)
jeffbolznv committed 3 days ago
Verified e321bdad
vulkan : one warp tile per subgroup in the small matmul tiles (#30106)
ianloic committed 3 days ago
Verified 459a1ba5
hexagon: generic dims collapse/reshape for unary, binary and matmul ops (#31071)
max-krasnyansky committed 3 days ago
Verified f0fb5984
opencl: add topk and lightning indexer (#31026)
lhez committed 3 days ago
Verified 9a4580b6
opencl: add initial q2_k support (#30999)
lhez committed 3 days ago
Verified b083c384
ci : update CPU job assignments (#31136)
ggerganov committed 3 days ago
Verified 344be0a6
chat : reject JSON nested more than 128 levels before walking it (#30242)
ykhrustalev committed 3 days ago
Verified b3fa6995
model: DSV4.1 flash (#31086)
am17an committed 3 days ago
Verified 74bca395
jinja : fix select and selectattr argument handling (#31059)
JoeyPatricio committed 4 days ago
Verified d327bf1f
hexagon: KV-head team split for multicore HMX flash attention (#31055)
ebateni committed 4 days ago
Verified 78a8ae14
server: refactor server entrypoint (#31113)
ngxson committed 4 days ago
Verified f3269ef9
Terms Of Use Privacy Policy Refund Policy Impressum