LLM inference in C/C++
C++ ggml
Commits on master
opencl: fold the gpt-oss MoE per-expert bias adds into the epilogue (op/kernel fusion) (#26431)
3af988f Hongqiang Wang committed 3 months ago
ci : fix empty release_id in make-release upload step (#27516)
a3b9c23 Georgi Gerganov committed 3 months ago
ci : run ccache-clear as the last step of release jobs (#27503)
d7fa69b Georgi Gerganov committed 3 months ago
scripts : add release.sh for release preparation (#27497)
c4b0225 Georgi Gerganov committed 3 months ago
Revert "sycl : add Q2_K reordered MMVQ and ESIMD kernels (#26336)" (#27486)
7a0e42f Georgi Gerganov committed 3 months ago
sycl : add Q2_K reordered MMVQ and ESIMD kernels (#26336)
ff14356 Todd Malsbary committed 3 months ago
test : make the FA V-is-view-of-K case a test case parameter (#27394)
5fff128 Georgi Gerganov committed 3 months ago
opencl: keep the vocab-scale K-quant lm_head on the CPU for Adreno A7X (compiler issue workaround) (#26440)
cd26896 Hongqiang Wang committed 3 months ago
sycl: Update gate logic for Alchemist GPUs regarding OneDNN features. (#26635)
1cb3f5e HumerousGorgon committed 3 months ago
sycl: fix multiple warnings in compiling sycl backend (#26713)
6602dd3 Ian Faust committed 3 months ago
ggml: support ggml_rope_set_offset on opencl, sycl, wgpu, hexagon (#27345)
749f688 Xuan-Son Nguyen committed 3 months ago