ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-03-12 06:50:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add abbility to merge up/gate expert tensors to Qwen3.5-MoE/Qwen3-Next (#1403)

Browse Source
This commit is contained in:
Kawrakow
2026-03-11 19:28:12 +01:00
committed by GitHub
parent 2161ee01cb
commit 1f4dcab5c6
1 changed files with 3 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/llama-load-tensors.cpp
View File

@@ -1378,7 +1378,7 @@ bool create_tensors_helper::create_qwen3next_tensors(const LLM_TN & tn) {
if (n_expert_used == 0) {
throw std::runtime_error("n_expert_used must be > 0 when QWEN3NEXT MoE tensors are present");
}
use_mmap_buffer &= !create_std_ffn_exps(n_embd, tn, i, llama_model_loader::TENSOR_NOT_REQUIRED, n_ff_exp);
use_mmap_buffer &= !create_std_ffn_exps(n_embd, tn, i, 0, n_ff_exp);
}
// Shared expert path (optional per-layer)
@@ -1450,9 +1450,7 @@ bool create_tensors_helper::create_qwen35moe_tensors(const LLM_TN & tn) {
}
layer.ffn_gate_inp = create_tensor(ctx_split, tn(LLM_TENSOR_FFN_GATE_INP, "weight", i), { n_embd, n_expert }, 0);
layer.ffn_gate_exps = create_tensor(ctx_split, tn(LLM_TENSOR_FFN_GATE_EXPS, "weight", i), { n_embd, n_ff_exp, n_expert }, 0);
layer.ffn_down_exps = create_tensor(ctx_split, tn(LLM_TENSOR_FFN_DOWN_EXPS, "weight", i), { n_ff_exp, n_embd, n_expert }, 0);
layer.ffn_up_exps = create_tensor(ctx_split, tn(LLM_TENSOR_FFN_UP_EXPS, "weight", i), { n_embd, n_ff_exp, n_expert }, 0);
use_mmap_buffer &= !create_std_ffn_exps(n_embd, tn, i, 0, n_ff_exp);
// Shared experts
const int64_t n_ff_shexp = hparams.n_ff_shexp ? hparams.n_ff_shexp : n_ff;
@@ -3137,7 +3135,7 @@ bool create_tensors_helper::merge_up_gate_exps(const LLM_TN & tn, int i, int bia
auto g_meta = ml.require_tensor_meta(g_name.c_str());
if (u_meta->type != g_meta->type || u_meta->ne[0] != g_meta->ne[0] || u_meta->ne[2] != g_meta->ne[2]) {
LLAMA_LOG_INFO("%s: not merging because up/fate meta info is different\n", __func__);
LLAMA_LOG_INFO("%s: not merging because up/gate meta info is different\n", __func__);
return false;
}