ggml-org/llama.cpp

Pull Requests Commits
convert : fix Nemotron-H LoRA GGUF conversion (#27356)
frozenblade1224 committed 59 days ago
Verified bcb6084a
ui: Replace per-conversation MCP overrides with per-conversation tool policy (#27745)
allozaur committed 59 days ago
Verified fe235f43
spec: Add benchmark-only synthetic speculative acceptance options (#27711)
gaugarg-nv committed 59 days ago
Verified 2bb9bdda
model : simplify MiniMax-01 graph (#27790)
fairydreaming committed 59 days ago
Verified deae5ee1
args: add --video-* CLI arguments (#24318)
ngxson committed 59 days ago
Verified f2955121
metal : fix memory leaks due to missing autoreleasepools (#27758)
nikwen committed 59 days ago
Verified 915dc6d3
llama : add --n-cpu-ffn option (#26622)
John-194 committed 59 days ago
Verified c5fc7e34
models : support nanbeige4.2-3B (#27730)
zqlcode committed 59 days ago
Verified d7a20741
hexagon: support for multi-NPU devices (IQ9, IQ10) and fully asynchronous backend (#26501)
max-krasnyansky committed 60 days ago
Verified 192067b7
llama: add token ID tracking to KV cell (#27762)
ngxson committed 60 days ago
Verified 925e1179
ui: Move Settings and MCP Servers routes to dialog-based views (#27744)
allozaur committed 60 days ago
Verified 539f2452
ui: Update Dialog component styling (#27743)
allozaur committed 60 days ago
Verified 0379a19f
vulkan: warptiles currently assume warp sizes <= 64, clamp to work around larger warps (#27726)
0cc4m committed 60 days ago
Verified 5e6a37cb
Implemented vulkan cross_entropy_loss and cross_entropy_loss_back (#27216)
PranavUttarkar committed 60 days ago
Verified bf942164
rpc : implement event and async backend APIs (#18626)
rgerganov committed 60 days ago
Verified d0132a68
ci: Clean up UI builds from releases (#27706)
allozaur committed 60 days ago
Verified 4d19b287
cuda: unblock mmq for MoE on sm_60 (#26264)
dfriehs committed 60 days ago
Verified fc35562b
ci : make cache bucket public (#27728)
CISC committed 60 days ago
Verified da9b5d68
conversion : fix Nemotron 3.5 Lightning layers (#27729)
danbev committed 60 days ago
Verified dac869b0
ggml-metal: add chunked SSD MMA for Mamba-2 prefill optimization (#26647)
dpantaleoni committed 60 days ago
Verified 11cd9884
ggml-meta: propagate buffer usage and call init on the new tensors (#27586)
max-krasnyansky committed 60 days ago
Verified 5d5cb4c3
kleidiai: Rework KleidiAI Build System/Integration (#26077)
JonathanC-ARM committed 61 days ago
Verified d222767c
ci : update OS used for ROCM to Ubuntu 24.04 (#27681)
superm1 committed 61 days ago
Verified eab8ee41
rpc: support apple RDMA as an RPC transport (#26421)
ryan5rdx committed 61 days ago
Verified b114b473
devops: use GGML_NATIVE=OFF for OpenVINO (#27338)
krzsztf committed 61 days ago
Verified 0a5ac49b
server : reject prefilled assistant messages with tool calls (#27626)
kyo-zzz committed 61 days ago
Verified 1729ed53
chat : scope qwen3-coder workarounds (#27679)
aldehir committed 61 days ago
Verified 0cc5b149
ci : store ccache on HF buckets (test with cuda-ubuntu for now) (#27699)
CISC committed 61 days ago
Verified 790b5713
ui: ESLint config updates (#27700)
allozaur committed 61 days ago
Verified f1357e49
metal : null-check buffer alloc to fix OOM crash (#25371)
ykhrustalev committed 61 days ago
Verified 3737e413
Terms Of Use Privacy Policy Refund Policy Impressum