From: forforever73 <redacted>
Date: Thu, 4 Jun 2026 13:56:33 +0000 (+0800)
Subject: return filter to save memory (#24125)
X-Git-Tag: upstream/0.0.10438~926
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=0dbfa66a1fb99e5270b8b6652f553c1bff59123f;p=pkg%2Fggml%2Fsources%2Fllama.cpp

return filter to save memory (#24125)

Co-authored-by: lvyichen <redacted>
---

diff --git a/src/llama-model.cpp b/src/llama-model.cpp
index 3c2a8e78b..bc7a83b15 100644
--- a/src/llama-model.cpp
+++ b/src/llama-model.cpp
@@ -2112,6 +2112,15 @@ llama_memory_i * llama_model::create_memory(const llama_memory_params & params,
                         filter = [n_main](int32_t il) { return (uint32_t)il >= n_main; };
                     }
 
+                    if (arch == LLM_ARCH_STEP35 && hparams.nextn_predict_layers > 0) {
+                        const uint32_t n_main = hparams.n_layer - hparams.nextn_predict_layers;
+                        if (params.ctx_type == LLAMA_CONTEXT_TYPE_MTP) {
+                            filter = [n_main](int32_t il) { return (uint32_t)il >= n_main; };
+                        } else {
+                            filter = [n_main](int32_t il) { return (uint32_t)il <  n_main; };
+                        }
+                    }
+
                     if (hparams.swa_type != LLAMA_SWA_TYPE_NONE) {
                         GGML_ASSERT(hparams.is_swa_any());