Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-19 01:05:09 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'upstream' into concedo_experimental

Browse Source 
# Conflicts:
#	.github/labeler.yml
#	.github/workflows/build-self-hosted.yml
#	benches/nemotron/nemotron-dgx-spark.md
#	docs/ops.md
#	docs/ops/SYCL.csv
#	ggml/src/ggml-cpu/kleidiai/kleidiai.cpp
#	ggml/src/ggml-sycl/backend.hpp
#	ggml/src/ggml-sycl/element_wise.cpp
#	ggml/src/ggml-sycl/element_wise.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	scripts/sync-ggml.last
#	tests/test-jinja.cpp
#	tests/test-llama-archs.cpp
This commit is contained in:
Concedo
2026-03-17 14:05:23 +08:00
parent 6d3f01d139 b6c83aad55
commit f31b040941
24 changed files with 1016 additions and 318 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+5 -1
View File
@@ -1701,6 +1701,7 @@ void llama_model::load_hparams(llama_model_loader & ml) {
}
} break;
case LLM_ARCH_DEEPSEEK2:
case LLM_ARCH_MISTRAL4:
{
// lite variants include DeepSeek-V2-Lite, GigaChat3-10B-A1.8B, Kanana-2-30B-A3B
const bool is_lite = (hparams.n_layer == 27 || hparams.n_layer == 26 || (hparams.n_layer == 48 && n_vocab == 128256));
@@ -5041,6 +5042,7 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
}
} break;
case LLM_ARCH_DEEPSEEK2:
case LLM_ARCH_MISTRAL4:
{
const bool is_mla = hparams.is_mla();
@@ -8008,7 +8010,7 @@ void llama_model::print_info() const {
LLAMA_LOG_INFO("%s: expert_weights_scale = %.1f\n", __func__, hparams.expert_weights_scale);
}
if (arch == LLM_ARCH_DEEPSEEK2 || arch == LLM_ARCH_GLM_DSA) {
if (arch == LLM_ARCH_DEEPSEEK2 || arch == LLM_ARCH_GLM_DSA || arch == LLM_ARCH_MISTRAL4) {
LLAMA_LOG_INFO("%s: n_layer_dense_lead = %d\n", __func__, hparams.n_layer_dense_lead);
LLAMA_LOG_INFO("%s: n_lora_q = %d\n", __func__, hparams.n_lora_q);
LLAMA_LOG_INFO("%s: n_lora_kv = %d\n", __func__, hparams.n_lora_kv);
@@ -8586,6 +8588,7 @@ ggml_cgraph * llama_model::build_graph(const llm_graph_params & params) const {
} break;
case LLM_ARCH_DEEPSEEK2:
case LLM_ARCH_GLM_DSA:
case LLM_ARCH_MISTRAL4:
{
llm = std::make_unique<llm_build_deepseek2>(*this, params);
} break;
@@ -8997,6 +9000,7 @@ llama_rope_type llama_model_rope_type(const llama_model * model) {
case LLM_ARCH_ERNIE4_5:
case LLM_ARCH_ERNIE4_5_MOE:
case LLM_ARCH_MISTRAL3:
case LLM_ARCH_MISTRAL4:
case LLM_ARCH_LLAMA_EMBED:
case LLM_ARCH_MAINCODER:
case LLM_ARCH_GLM_DSA: