From 69c9f199574e7fe3a05847f6652bf914639f4c08 Mon Sep 17 00:00:00 2001
From: Harshal Janjani <harshaljanjani@gmail.com>
Date: Fri, 29 May 2026 01:48:12 +0400
Subject: [PATCH] fix(frontend): Add multimodal placeholders to Gemma4 tool
 message template (#41459)

Signed-off-by: Harshal Janjani <harshaljanjani@gmail.com>
Co-authored-by: Ben Browning <bbrownin@redhat.com>
---
 examples/tool_chat_template_gemma4.jinja     |  9 +++
 tests/renderers/test_gemma4_chat_template.py | 69 ++++++++++++++++++++
 2 files changed, 78 insertions(+)

diff --git a/examples/tool_chat_template_gemma4.jinja b/examples/tool_chat_template_gemma4.jinja
index d61dd795b5..ef76582310 100644
--- a/examples/tool_chat_template_gemma4.jinja
+++ b/examples/tool_chat_template_gemma4.jinja
@@ -295,6 +295,15 @@
                                 {%- endif -%}
                             {%- endfor -%}
                             {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
+                            {%- for part in tool_body -%}
+                                {%- if part.get('type') == 'image' -%}
+                                    {{- '<|image|>' -}}
+                                {%- elif part.get('type') == 'audio' -%}
+                                    {{- '<|audio|>' -}}
+                                {%- elif part.get('type') == 'video' -%}
+                                    {{- '<|video|>' -}}
+                                {%- endif -%}
+                            {%- endfor -%}
                         {%- else -%}
                             {{- format_tool_response_block(ns_tname.name, tool_body) -}}
                         {%- endif -%}
diff --git a/tests/renderers/test_gemma4_chat_template.py b/tests/renderers/test_gemma4_chat_template.py
index a4a0b41d05..ac13c0d4d5 100644
--- a/tests/renderers/test_gemma4_chat_template.py
+++ b/tests/renderers/test_gemma4_chat_template.py
@@ -343,3 +343,72 @@ class TestGemma4ChatTemplate:
         assert '<|"|>Alice<|"|>' in result
         assert "active:true" in result
         assert "count:42" in result
+
+    def test_tool_response_with_multimodal_content(self, gemma4_template):
+        """Multimodal placeholders in tool messages are emitted after the
+        tool_response block."""
+        messages = [
+            {"role": "user", "content": "Download the image and describe it."},
+            {
+                "role": "assistant",
+                "content": "",
+                "tool_calls": [
+                    {
+                        "id": "call_1",
+                        "type": "function",
+                        "function": {
+                            "name": "download_image",
+                            "arguments": '{"url": "https://example.com/x.png"}',
+                        },
+                    },
+                ],
+            },
+            {
+                "role": "tool",
+                "tool_call_id": "call_1",
+                "content": [
+                    {"type": "text", "text": "Image downloaded successfully."},
+                    {"type": "image"},
+                ],
+            },
+        ]
+        result = _render(gemma4_template, messages, add_generation_prompt=True)
+        assert "<|tool_response>" in result
+        assert "response:download_image{" in result
+        assert "<tool_response|>" in result
+        assert "<|image|>" in result
+
+    def test_tool_response_with_all_modalities(self, gemma4_template):
+        """All multimodal types (image, audio, video) in a single tool
+        response are rendered."""
+        messages = [
+            {"role": "user", "content": "Process media"},
+            {
+                "role": "assistant",
+                "content": "",
+                "tool_calls": [
+                    {
+                        "id": "c1",
+                        "type": "function",
+                        "function": {
+                            "name": "process",
+                            "arguments": "{}",
+                        },
+                    },
+                ],
+            },
+            {
+                "role": "tool",
+                "tool_call_id": "c1",
+                "content": [
+                    {"type": "text", "text": "Results."},
+                    {"type": "image"},
+                    {"type": "audio"},
+                    {"type": "video"},
+                ],
+            },
+        ]
+        result = _render(gemma4_template, messages, add_generation_prompt=True)
+        assert "<|image|>" in result
+        assert "<|audio|>" in result
+        assert "<|video|>" in result