From a1cb5b5dc05d2540640069b9549dd93557c81a16 Mon Sep 17 00:00:00 2001
From: oobabooga <112222186+oobabooga@users.noreply.github.com>
Date: Thu, 2 Apr 2026 21:56:06 -0700
Subject: [PATCH] llama.cpp: Disable jinja by default (we use Python jinja, not
 cpp jinja)

This was causing template compilation issues with qwen models.
---
 modules/llama_cpp_server.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/modules/llama_cpp_server.py b/modules/llama_cpp_server.py
index 2d873f00..a4390adb 100644
--- a/modules/llama_cpp_server.py
+++ b/modules/llama_cpp_server.py
@@ -418,6 +418,7 @@ class LlamaServer:
             "--ubatch-size", str(shared.args.ubatch_size),
             "--port", str(self.port),
             "--no-webui",
+            "--no-jinja",
             "--flash-attn", "on",
         ]