Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-08 05:49:10 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'upstream' into concedo_experimental

Browse Source 
# Conflicts:
#	.github/workflows/release.yml
#	.github/workflows/ui-build-self-hosted.yml
#	.github/workflows/ui-build.yml
#	.github/workflows/ui-publish.yml
#	.github/workflows/ui-self-hosted.yml
#	.github/workflows/ui.yml
#	.gitignore
#	README.md
#	docs/ops.md
#	docs/ops/Vulkan.csv
#	ggml/CMakeLists.txt
#	scripts/sync-ggml.last
#	scripts/sync_vendor.py
#	scripts/ui-assets.cmake
#	tests/test-jinja.cpp
#	tests/test-llama-archs.cpp
This commit is contained in:
Concedo
2026-06-14 11:26:16 +08:00
parent ef5a724ee4 c2ba3e47a2
commit ea0351c71a
117 changed files with 11026 additions and 2390 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+9 -1
View File
@@ -52,6 +52,7 @@
#include "models/codeshell.cpp"
#include "models/cogvlm.cpp"
#include "models/cohere2.cpp"
#include "models/cohere2moe.cpp"
#include "models/command-r.cpp"
#include "models/dbrx.cpp"
#include "models/deci.cpp"
@@ -292,6 +293,8 @@ static llama_model * llama_model_mapping(llm_arch arch, const llama_model_params
return new llama_model_command_r(params);
case LLM_ARCH_COHERE2:
return new llama_model_cohere2(params);
case LLM_ARCH_COHERE2MOE:
return new llama_model_cohere2moe(params);
case LLM_ARCH_DBRX:
return new llama_model_dbrx(params);
case LLM_ARCH_OLMO:
@@ -1602,9 +1605,12 @@ bool llama_model_base::load_tensors(llama_model_loader & ml) {
}
ml.done_getting_tensors();
// Tied NVFP4 output is valid when no separate LM-head scale tensors are present.
// If sidecar scales exist, the output weight must be an actual output tensor.
GGML_ASSERT(!(output && tok_embd &&
strcmp(output->name, tok_embd->name) == 0 &&
output->type == GGML_TYPE_NVFP4));
output->type == GGML_TYPE_NVFP4 &&
(output_s || output_in_s)));
// populate tensors_by_name
for (auto & [_, ctx_ptr] : ml.ctx_map) {
for (auto * cur = ggml_get_first_tensor(ctx_ptr.get()); cur != NULL; cur = ggml_get_next_tensor(ctx_ptr.get(), cur)) {
@@ -1979,6 +1985,7 @@ void llama_model::print_info() const {
}
if (arch == LLM_ARCH_MELLUM ||
arch == LLM_ARCH_COHERE2MOE ||
arch == LLM_ARCH_QWEN3MOE ||
arch == LLM_ARCH_OPENAI_MOE ||
arch == LLM_ARCH_QWEN3VLMOE ||
@@ -2524,6 +2531,7 @@ llama_rope_type llama_model_rope_type(const llama_model * model) {
case LLM_ARCH_XVERSE:
case LLM_ARCH_COMMAND_R:
case LLM_ARCH_COHERE2:
case LLM_ARCH_COHERE2MOE:
case LLM_ARCH_OLMO:
case LLM_ARCH_ARCTIC:
case LLM_ARCH_DEEPSEEK: