common : add --no-warmup option for main/llama-cli (#8712)

This commit adds a --no-warmup option for llama-cli.

The motivation for this is that it can be convenient to skip the
warmup llama_decode call when debugging.

Signed-off-by: Daniel Bevenius <redacted>

diff --git a/common/common.cpp b/common/common.cpp
index ec44a05521c9d7379779239a03f09da734b77951..60c7eac75c613bfccf3002ae93b2b57de6970ac8 100644 (file)
--- a/common/common.cpp
+++ b/common/common.cpp
@@ -1324,6 +1324,10 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
         else { invalid_param = true; }
         return true;
     }
+    if (arg == "--no-warmup") {
+        params.warmup = false;
+        return true;
+    }
 #ifndef LOG_DISABLE_LOGS
     // Parse args for logging parameters
     if (log_param_single_parse(argv[i])) {
@@ -1446,6 +1450,7 @@ void gpt_params_print_usage(int /*argc*/, char ** argv, const gpt_params & param
     options.push_back({ "main infill", "       --in-prefix-bos",        "prefix BOS to user inputs, preceding the `--in-prefix` string" });
     options.push_back({ "main infill", "       --in-prefix STRING",     "string to prefix user inputs with (default: empty)" });
     options.push_back({ "main infill", "       --in-suffix STRING",     "string to suffix after user inputs with (default: empty)" });
+    options.push_back({ "main",        "       --no-warmup",            "skip warming up the model with an empty run" });
     options.push_back({ "server infill",
                                        "       --spm-infill",           "use Suffix/Prefix/Middle pattern for infill (instead of Prefix/Suffix/Middle) as some models prefer this. (default: %s)", params.spm_infill ? "enabled" : "disabled" });