Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

common, server : enable preserve_reasoning kwarg by default, log its effective state (#28174)

Browse Source 
* common, server : enable preserve_reasoning kwarg by default, log its effective state

If the preserve_reasoning chat template kwarg is not specified explicitly
via --reasoning-preserve / --no-reasoning-preserve, it is enabled by
default after argument processing. The server logs the effective state of
the kwarg, warns that it is enabled by default when the template supports
it, and only warns "has no effect" when it was enabled explicitly on a
template that does not support it. Setting the kwarg via
--chat-template-kwargs is deprecated.

Assisted-by: pi:llama.cpp/Qwen3.8-27B

* cont : update comment

Co-authored-by: Xuan-Son Nguyen <son@huggingface.co>

---------

Co-authored-by: Xuan-Son Nguyen <son@huggingface.co>
This commit is contained in:
Georgi Gerganov
2026-09-02 19:19:54 +03:00
committed by GitHub
co-authored by Xuan-Son Nguyen
parent 7798007a29
commit e750b887a8
3 changed files with 26 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-context.cpp
+13 -2
View File
@@ -1493,11 +1493,22 @@ private:
auto caps = common_chat_templates_get_caps(chat_params.tmpls.get());
auto it = params_base.default_template_kwargs.find("preserve_reasoning");
bool supported = caps.at("supports_preserve_reasoning");
bool enabled = it != params_base.default_template_kwargs.end();
bool specified = params_base.preserve_reasoning_specified;
// note: the kwarg is enabled by default if not specified explicitly, so check the value
bool enabled = it != params_base.default_template_kwargs.end() && it->second == "true";
if (supported) {
SRV_TRC("preserve_reasoning kwarg: %s\n",
it == params_base.default_template_kwargs.end() ? "unset (template default)" : it->second.c_str());
} else {
SRV_TRC("%s", "preserve_reasoning kwarg: not supported by template\n");
}
if (supported && !specified) {
SRV_WRN("%s", "chat template supports preserving reasoning, it is enabled by default (may use more tokens, disable via --no-reasoning-preserve)\n");
}
if (supported && !enabled) {
SRV_INF("%s", "chat template supports preserving reasoning, consider enabling it via --reasoning-preserve\n");
}
if (!supported && enabled) {
if (!supported && specified && enabled) {
SRV_WRN("%s", "chat template does NOT support preserving reasoning, --reasoning-preserve has no effect\n");
}
}