talk-llama : sync llama.cpp

This commit is contained in:
Georgi Gerganov
2026-06-26 16:03:57 +03:00
parent 0279b5386a
commit 0ae02cdb2c
17 changed files with 101 additions and 59 deletions
@@ -169,7 +169,6 @@ ggml_tensor * llm_build_mamba_base::build_mamba2_layer(llm_graph_input_rs * inp,
GGML_ASSERT(ubatch.equal_seqs());
GGML_ASSERT(ubatch.n_tokens == n_seq_tokens * n_seqs);
GGML_ASSERT(d_inner % n_head == 0);
GGML_ASSERT(d_inner % d_state == 0);
GGML_ASSERT(d_inner % n_group == 0);
ggml_tensor * conv_states_all = mctx_cur->get_r_l(il);