vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-05-09 19:46:11 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	ggml/src/ggml-sycl/gemm.hpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/set_rows.cpp
This commit is contained in:
Concedo 2025-07-14 23:16:44 +08:00
commit 4db8ba6228
5 changed files with 47 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/llama-chat.cpp
View file

@ -170,7 +170,7 @@ llm_chat_template llm_chat_detect_template(const std::string & tmpl) {
// ref: https://huggingface.co/LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct/discussions/8#66bae61b1893d14ee8ed85bb
// EXAONE-3.0-7.8B-Instruct
return LLM_CHAT_TEMPLATE_EXAONE_3;
} else if (tmpl_contains("rwkv-world")) {
} else if (tmpl_contains("rwkv-world") || tmpl_contains("{{- 'User: ' + message['content']|trim + '\\n\\n' -}}")) {
return LLM_CHAT_TEMPLATE_RWKV_WORLD;
} else if (tmpl_contains("<|start_of_role|>")) {
return LLM_CHAT_TEMPLATE_GRANITE;

3
src/llama-quant.cpp
View file

@ -887,8 +887,7 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
if (std::regex pattern(tname); std::regex_search(tensor_name, pattern)) {
if (qtype != new_type) {
LLAMA_LOG_DEBUG("(overriding %s) ", ggml_type_name(new_type));
new_type = qtype;
break; // if two or more types are specified for the tensor, first match wins
new_type = qtype; // if two or more types are specified for the same tensor, the last match wins
}
}
}