Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-18 16:55:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'upstream' into concedo_experimental

Browse Source 
# Conflicts:
#	.github/workflows/check-vendor.yml
#	.github/workflows/close-issue.yml
#	.github/workflows/editorconfig.yml
#	.github/workflows/gguf-publish.yml
#	.github/workflows/labeler.yml
#	.github/workflows/pre-tokenizer-hashes.yml
#	.github/workflows/python-check-requirements.yml
#	.github/workflows/python-lint.yml
#	.github/workflows/python-type-check.yml
#	.github/workflows/server.yml
#	.github/workflows/update-ops-docs.yml
#	README.md
#	docs/build.md
#	examples/model-conversion/scripts/utils/perplexity-gen.sh
#	examples/model-conversion/scripts/utils/perplexity-run-simple.sh
#	examples/model-conversion/scripts/utils/perplexity-run.sh
#	examples/model-conversion/scripts/utils/quantize.sh
#	examples/model-conversion/scripts/utils/run-embedding-server.sh
#	ggml/src/ggml-cpu/ggml-cpu.c
#	ggml/src/ggml-hexagon/htp/flash-attn-ops.c
#	ggml/src/ggml-opencl/CMakeLists.txt
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-opencl/kernels/cvt.cl
#	ggml/src/ggml-opencl/kernels/mul_mv_q6_k_f32.cl
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	scripts/compare-llama-bench.py
#	tests/test-backend-ops.cpp
#	tests/test-gguf.cpp
#	tools/cli/README.md
#	tools/completion/README.md
#	tools/server/README.md
This commit is contained in:
Concedo
2026-01-27 23:06:13 +08:00
parent 3a84623ea5 68ac3acb43
commit f6ece6fd37
40 changed files with 3108 additions and 477 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common/chat.cpp
+51
View File
@@ -2664,6 +2664,51 @@ static common_chat_params common_chat_params_init_exaone_moe(const common_chat_t
return data;
}
static common_chat_params common_chat_params_init_translate_gemma(const common_chat_template & tmpl, const struct templates_params & inputs) {
common_chat_params data;
// This template does not support tools or reasoning
// we just need to transform the messages into the correct schema
templates_params inputs_new = inputs;
json & messages = inputs_new.messages;
// default to chat_template_kwargs, or en-GB if not specified
std::string default_src_lang = inputs.extra_context.value("source_lang_code", "en-GB");
std::string default_tgt_lang = inputs.extra_context.value("target_lang_code", "en-GB");
GGML_ASSERT(messages.is_array());
for (auto & message : messages) {
if (message.contains("role") && message["role"].get<std::string>() != "user") {
continue;
}
if (!message.contains("content")) {
message["content"] = json::array();
}
if (message.contains("content") && !message["content"].is_array()) {
auto content_str = message["content"].get<std::string>();
// default to en-GB if not specified (to make common_chat_format_example works)
auto src_lang = message.contains("source_lang_code")
? message["source_lang_code"].get<std::string>() : default_src_lang;
auto tgt_lang = message.contains("target_lang_code")
? message["target_lang_code"].get<std::string>() : default_tgt_lang;
message["content"] = json::array({
json{
{"type", "text"},
{"text", content_str},
{"source_lang_code", src_lang},
{"target_lang_code", tgt_lang},
}
});
}
}
data.prompt = apply(tmpl, inputs_new, std::nullopt, std::nullopt);
data.format = COMMON_CHAT_FORMAT_GENERIC;
return data;
}
static common_chat_params common_chat_params_init_without_tools(const common_chat_template & tmpl, const struct templates_params & inputs) {
common_chat_params data;
data.prompt = apply(tmpl, inputs);
@@ -3059,6 +3104,12 @@ static common_chat_params common_chat_templates_apply_jinja(
return common_chat_params_init_solar_open(tmpl, params);
}
// TranslateGemma
if (src.find("[source_lang_code]") != std::string::npos &&
src.find("[target_lang_code]") != std::string::npos) {
return common_chat_params_init_translate_gemma(tmpl, params);
}
// Plain handler (no tools)
if (params.tools.is_null() || inputs.tool_choice == COMMON_CHAT_TOOL_CHOICE_NONE) {
return common_chat_params_init_without_tools(tmpl, params);