llama.cpp
e95dae18 - Remove padding and multiple D2D copies for MTP (#24086)

Commit
61 days ago
Remove padding and multiple D2D copies for MTP (#24086) * Make ggml_gated_delta_net take only the initial recurrent state (D, 1, n_seqs) and passes the snapshot count K as an op parameter instead of inferring it from state->ne[1]. Remove the padding hack and copy all emitted snapshots into the recurrent cache with a single strided ggml_cpy * Make GDN changes in all backends. Address review comments. * Fix CI build errors
Author
Parents
Loading