LLM inference in C/C++
C++ ggml
Commits on master
sycl: fuse the gated-delta-net state writeback cpy (#26643)
3d93885 Titaniumtown committed 3 months ago
dflash : clarify output logging of target_layer_ids (#27013)
2bacf9e Daniel Bevenius committed 3 months ago
ui: Refactor data-attrs constants, enum for bool strings (#27002)
bdffafa Aleksander Grygier committed 3 months ago
OpenVINO: Qwen3.5, memory optimization, and test-recurrent-state-rollback (#26952)
aee56b3 Zijun Yu committed 3 months ago
[SYCL] Support host pinned mem to improve SYCL Host-to-Device Memory Access (#26789)
a97123e Neo Zhang committed 3 months ago
chat : fix LFM2 tool call arg name prefix ambiguity (#26960)
2606220 Aldehir Rojas committed 3 months ago
common : auto-detect spec type from draft GGUF metadata (#26814)
f65e568 aic0d3r committed 3 months ago
spec: enable backend sampling for both dflash & dspark (#26958)
0d0bfcd Ruixiang Wang committed 3 months ago
ggml-cpu/ops: vectorize flash-attention V-cache F16 to F32 conversion (#26947)
eeae28b jinzihao committed 3 months ago
sycl: remove separate fp32 type promotion in gemm non-oneDNN path (#26372)
154d57a Ian Faust committed 3 months ago
sycl : enhance concat to support Q4_0, Q4_1, Q5_0, Q5_1, Q8_0 (#26800)
d415e65 Neo Zhang committed 3 months ago
server: refactor + correctness fixes for metrics (#26920)
decaf50 Xuan-Son Nguyen committed 3 months ago
ui: Clean up contexts, remove prop drilling from Chat Form Actions (#26951)
d86c7d6 Aleksander Grygier committed 3 months ago
common: add system-level config file (#26118)
8e7f22b Johnathan Craig Maudlin committed 3 months ago