[model] support Mistral3.1 small 2503 (#8335)

src/llamafactory/data/mm_plugin.py

@@ -1274,9 +1274,10 @@ class PixtralPlugin(BasePlugin):
             content = message["content"]
             while IMAGE_PLACEHOLDER in content:
                 if self.expand_mm_tokens:
+                    patch_size = processor.patch_size * getattr(processor, "spatial_merge_size", 1)
                     height, width = next(image_sizes)
-                    num_height_tokens = height // processor.patch_size
-                    num_width_tokens = width // processor.patch_size
+                    num_height_tokens = height // patch_size
+                    num_width_tokens = width // patch_size
                     replace_tokens = [[self.image_token] * num_width_tokens + [image_break_token]] * num_height_tokens
                     replace_tokens = [item for sublist in replace_tokens for item in sublist]  # flatten list
                     replace_tokens[-1] = image_end_token

src/llamafactory/data/template.py

@@ -1433,6 +1433,7 @@ register_template(
     format_observation=StringFormatter(slots=["""[TOOL_RESULTS]{"content": {{content}}}[/TOOL_RESULTS]"""]),
     format_tools=ToolFormatter(tool_format="mistral"),
     format_prefix=EmptyFormatter(slots=[{"bos_token"}]),
+    mm_plugin=get_mm_plugin(name="pixtral", image_token="[IMG]"),
 )