diff --git a/app/src/protyle/wysiwyg/enter.ts b/app/src/protyle/wysiwyg/enter.ts
index 403c66381..4244b17ad 100644
--- a/app/src/protyle/wysiwyg/enter.ts
+++ b/app/src/protyle/wysiwyg/enter.ts
@@ -296,7 +296,6 @@ export const enter = (blockElement: HTMLElement, range: Range, protyle: IProtyle
 
 const listEnter = (protyle: IProtyle, blockElement: HTMLElement, range: Range) => {
     const listItemElement = blockElement.parentElement;
-
     const editableElement = getContenteditableElement(blockElement);
     if (// \n 是因为 https://github.com/siyuan-note/siyuan/issues/3846
         ["", "\n"].includes(editableElement.textContent) &&
@@ -451,6 +450,10 @@ const listEnter = (protyle: IProtyle, blockElement: HTMLElement, range: Range) =
         !hasNextSibling(editableElement?.lastElementChild)) {
         editableElement.insertAdjacentText("beforeend", "\n");
     }
+    // img 后有文字，在 img 后换行
+    if (editableElement?.lastElementChild?.classList.contains("img") && !hasNextSibling(editableElement?.lastElementChild)) {
+        editableElement.insertAdjacentText("beforeend", Constants.ZWSP);
+    }
     getContenteditableElement(newElement).appendChild(selectNode);
     listItemElement.insertAdjacentElement("afterend", newElement);
     if (listItemElement.getAttribute("data-subtype") === "o") {
diff --git a/app/src/protyle/wysiwyg/index.ts b/app/src/protyle/wysiwyg/index.ts
index b06a9d577..9641ca60a 100644
--- a/app/src/protyle/wysiwyg/index.ts
+++ b/app/src/protyle/wysiwyg/index.ts
@@ -2400,12 +2400,19 @@ export class WYSIWYG {
             const imgElement = hasTopClosestByClassName(event.target, "img");
             if (!event.shiftKey && !ctrlIsPressed && imgElement) {
                 imgElement.classList.add("img--select");
-                range.setStartAfter(imgElement);
-                range.collapse(true);
-                focusByRange(range);
-                // 需等待 range 更新再次进行渲染
-                if (protyle.options.render.breadcrumb) {
-                    protyle.breadcrumb.render(protyle);
+                const nextSibling = hasNextSibling(imgElement)
+                if (nextSibling) {
+                    if (nextSibling.textContent.startsWith(Constants.ZWSP)) {
+                        range.setStart(nextSibling, 1);
+                    } else {
+                        range.setStart(nextSibling, 0);
+                    }
+                    range.collapse(true);
+                    focusByRange(range);
+                    // 需等待 range 更新再次进行渲染
+                    if (protyle.options.render.breadcrumb) {
+                        protyle.breadcrumb.render(protyle);
+                    }
                 }
                 return;
             }