Fix apply_chat_template crash on tool_call messages without content

src/transformers/cli/serving/utils.py

@@ -930,24 +930,25 @@ def get_processor_inputs_from_messages(messages: list[dict], modality: Modality)
         for message in messages:
             parsed = {"role": message["role"], "content": []}
 
+            content = message.get("content")
             if modality == Modality.LLM:
-                if isinstance(message["content"], str):
-                    parsed["content"] = message["content"]
-                elif isinstance(message["content"], list):
-                    texts = [c["text"] for c in message["content"] if c["type"] == "text"]
+                if isinstance(content, str):
+                    parsed["content"] = content
+                elif isinstance(content, list):
+                    texts = [c["text"] for c in content if c["type"] == "text"]
                     parsed["content"] = " ".join(texts)
 
             elif modality == Modality.VLM:
-                if isinstance(message["content"], str):
-                    parsed["content"].append({"type": "text", "text": message["content"]})
-                else:
-                    for content in message["content"]:
-                        if content["type"] == "text":
-                            parsed["content"].append(content)
-                        elif content["type"] == "image_url":
+                if isinstance(content, str):
+                    parsed["content"].append({"type": "text", "text": content})
+                elif isinstance(content, list):
+                    for content_block in content:
+                        if content_block["type"] == "text":
+                            parsed["content"].append(content_block)
+                        elif content_block["type"] == "image_url":
                             from PIL import Image
 
-                            url = content["image_url"]["url"]
+                            url = content_block["image_url"]["url"]
                             if "base64" in url:
                                 image_data = re.sub("^data:image/.+;base64,", "", url)
                                 image = Image.open(BytesIO(base64.b64decode(image_data)))

src/transformers/models/smolvlm/processing_smolvlm.py

@@ -333,7 +333,7 @@ def apply_chat_template(
             (isinstance(content, dict) and content["type"] == "video")
             for conversation in conversations
             for message in conversation
-            for content in message["content"]
+            for content in (message.get("content") or [])
         )
         if chat_template is None and has_video:
             # re-assign to the correct default template for BC, if user is not requesting their own template

src/transformers/processing_utils.py

@@ -1804,12 +1804,15 @@ def apply_chat_template(
             for conversation in conversations:
                 images, videos = [], []
                 for message in conversation:
-                    visuals = [content for content in message["content"] if content["type"] in ["image", "video"]]
+                    content = message.get("content") or []
+                    visuals = [
+                        content_block for content_block in content if content_block["type"] in ["image", "video"]
+                    ]
                     audio_fnames = [
-                        content[key]
-                        for content in message["content"]
+                        content_block[key]
+                        for content_block in content
                         for key in ["audio", "url", "path"]
-                        if key in content and content["type"] == "audio"
+                        if key in content_block and content_block["type"] == "audio"
                     ]
                     image_fnames = [
                         vision_info[key]

src/transformers/utils/chat_template_utils.py

@@ -543,8 +543,10 @@ def render_jinja_template(
             chat = chat.messages
         if continue_final_message:
             chat = deepcopy(chat)
-            final_message = chat[-1]["content"]
-            if isinstance(final_message, (list, tuple)):
+            final_message = chat[-1].get("content")
+            if final_message is None:
+                raise ValueError("continue_final_message is set but the final message has no content to continue!")
+            elif isinstance(final_message, (list, tuple)):
                 for content_block in reversed(final_message):
                     if "text" in content_block:
                         # Pick the last text block in the message (the first one we hit while iterating in reverse)

tests/test_processing_common.py

@@ -1990,6 +1990,31 @@ def test_apply_chat_template_assistant_mask(self):
         ids_is_same = processor.tokenizer.encode(assistant_text, add_special_tokens=False), assistant_ids.tolist()
         self.assertTrue(text_is_same or ids_is_same)
 
+    @require_torch
+    def test_apply_chat_template_tool_calls_no_content(self):
+        processor = self.get_processor()
+
+        if processor.chat_template is None:
+            self.skipTest("Processor has no chat template")
+
+        if "tool" not in processor.chat_template:  # good heuristic to check if template supports tools
+            self.skipTest("Chat template does not support tools")
+
+        messages = [
+            {
+                "role": "user",
+                "content": [{"type": "text", "text": "What is the weather?"}],
+            },
+            {
+                "role": "assistant",
+                "tool_calls": [{"type": "function", "function": {"name": "get_weather", "arguments": "{}"}}],
+            },
+        ]
+
+        # Regression test for #45290: tokenize=True used to raise KeyError when "content" was missing
+        result = processor.apply_chat_template(messages, tokenize=True)
+        self.assertIsInstance(result, list)
+
     def test_get_num_multimodal_tokens_matches_processor_call(self):
         "Tests that the helper used internally in vLLM works correctly"