LLM inference in C/C++
C++ ggml
models : remove duplicate metadata load (#27378)
Sigbjørn Skjæret2e92ecd0247d
diff --git a/src/models/deepseek32.cpp b/src/models/deepseek32.cppindex 08555a8016..2b82a780c4 100644--- a/src/models/deepseek32.cpp+++ b/src/models/deepseek32.cpp@@ -10,8 +10,6 @@ void llama_model_deepseek32::load_arch_hparams(llama_model_loader & ml) {ml.get_key_or_arr(LLM_KV_ROPE_DIMENSION_SECTIONS, hparams.rope_sections, 4, false);// MoE parameters- ml.get_key(LLM_KV_EXPERT_COUNT, hparams.n_expert);- ml.get_key(LLM_KV_EXPERT_USED_COUNT, hparams.n_expert_used);ml.get_key(LLM_KV_EXPERT_SHARED_COUNT, hparams.n_expert_shared);ml.get_key(LLM_KV_LEADING_DENSE_BLOCK_COUNT, hparams.n_layer_dense_lead, false);ml.get_key(LLM_KV_EXPERT_WEIGHTS_SCALE, hparams.expert_weights_scale, false);diff --git a/src/models/glm-dsa.cpp b/src/models/glm-dsa.cppindex 803ef76747..93a1448b46 100644--- a/src/models/glm-dsa.cpp+++ b/src/models/glm-dsa.cpp@@ -32,8 +32,6 @@ void llama_model_glm_dsa::load_arch_hparams(llama_model_loader & ml) {ml.get_key_or_arr(LLM_KV_ROPE_DIMENSION_SECTIONS, hparams.rope_sections, 4, false);// MoE parameters- ml.get_key(LLM_KV_EXPERT_COUNT, hparams.n_expert);- ml.get_key(LLM_KV_EXPERT_USED_COUNT, hparams.n_expert_used);ml.get_key(LLM_KV_EXPERT_SHARED_COUNT, hparams.n_expert_shared);ml.get_key(LLM_KV_LEADING_DENSE_BLOCK_COUNT, hparams.n_layer_dense_lead, false);ml.get_key(LLM_KV_EXPERT_WEIGHTS_SCALE, hparams.expert_weights_scale, false);diff --git a/src/models/glm4-moe.cpp b/src/models/glm4-moe.cppindex d60e47ddf0..8cde66978c 100644--- a/src/models/glm4-moe.cpp+++ b/src/models/glm4-moe.cpp@@ -6,8 +6,6 @@ void llama_model_glm4_moe::load_arch_hparams(llama_model_loader & ml) {ml.get_key_or_arr(LLM_KV_ROPE_DIMENSION_SECTIONS, hparams.rope_sections, 4, false);// MoE parameters- ml.get_key(LLM_KV_EXPERT_COUNT, hparams.n_expert);- ml.get_key(LLM_KV_EXPERT_USED_COUNT, hparams.n_expert_used);ml.get_key(LLM_KV_EXPERT_SHARED_COUNT, hparams.n_expert_shared);ml.get_key(LLM_KV_LEADING_DENSE_BLOCK_COUNT, hparams.n_layer_dense_lead, false);ml.get_key(LLM_KV_EXPERT_WEIGHTS_SCALE, hparams.expert_weights_scale, false);