LLM inference in C/C++
C++ ggml
Commits on master
server : fix dead LLAMA_ARG_HF_REPO_FILE key in preset allow-list (#29938)
bf9a0cc Adrien Gallouët committed 1 month ago
mtmd : fix deprecated strdup warning on Windows (#29863)
836d571 Adrien Gallouët committed 1 month ago
vendor : update cpp-httplib to 0.59.0 (#29886)
eec18f5 Alessandro de Oliveira Faria (A.K.A.CABELO) committed 1 month ago
server : fix laya abort by limiting n_batch to n_ubatch (#29903)
1537a0a Nik Bogatyrev committed 1 month ago
common : add common_is_tty() helper and fix deprecated warnings on Windows (#29860)
edd6e2b Adrien Gallouët committed 1 month ago
ci: fix flaky ADD_ADD f16 by using the fused ADD tolerance (#29904)
a55e952 Pascal committed 1 month ago
graph: gather the recurrent states once so the reserve covers every split (#29856)
436f6f8 Pascal committed 1 month ago
ggml-openvino: update to 2026.4.1, optimize performance, expand ops, improve device listing. (#29852)
b92761a Ravi Panchumarthy committed 1 month ago
model : Add LFM2.5-Encoder-350M and LFM2.5-Encoder-230M (#29862)
cb7934c Tarek Dakhran committed 1 month ago
model: add support for clef decision model (text-only) (#29831)
99b9548 Xuan-Son Nguyen committed 1 month ago
spec : add probabilistic sampling for simple draft and MTP (#27694)
1fb7ef3 Pranesh Gonegandla committed 1 month ago
ggml-cuda : fix cpy transposed path corrupting non-contiguous dst (#27663)
134b2bb Yash Raj Pandey committed 1 month ago
ggml-quants : avoid invalid rounding in qkx3 scale search (#29817)
2923cf2 Yash Raj Pandey committed 1 month ago
ggml-cpu : fix soft_max_back wrong output when dst aliases src1 (#27096)
dd4c286 Yash Raj Pandey committed 1 month ago
metal : add tensor API flash attention kernel for F16 KV (#29570)
926862e Ethan Guo committed 1 month ago
llama, server: add /v1/systemone API (models: laya, julia-1, lev, openjev, kev) (#29818)
a4cb4c6 Xuan-Son Nguyen committed 1 month ago
common : remove fs_open_ifstream() by using u8path() (#29841)
70849ee Adrien Gallouët committed 1 month ago