From 919fe432c3961de9c001cce25d6083b4bfef6f2f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?DAN=E2=84=A2?= <dranger003@gmail.com>
Date: Tue, 17 Dec 2024 11:02:26 -0500
Subject: [PATCH] Bump model_template to 16384 bytes to support larger chat
 templates.

---
 src/llama.cpp | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/llama.cpp b/src/llama.cpp
index 1cc8a9332..cae86124b 100644
--- a/src/llama.cpp
+++ b/src/llama.cpp
@@ -22652,7 +22652,7 @@ int32_t llama_chat_apply_template(
     if (tmpl == nullptr) {
         GGML_ASSERT(model != nullptr);
         // load template from model
-        std::vector<char> model_template(2048, 0); // longest known template is about 1200 bytes
+        std::vector<char> model_template(16384, 0); // longest known template is about 12792 bytes
         std::string template_key = "tokenizer.chat_template";
         int32_t res = llama_model_meta_val_str(model, template_key.c_str(), model_template.data(), model_template.size());
         if (res < 0) {