diff --git a/core/pom.xml b/core/pom.xml
index d3235cb18..273e314fb 100644
--- a/core/pom.xml
+++ b/core/pom.xml
@@ -39,11 +39,6 @@
       <groupId>com.google.guava</groupId>
       <artifactId>guava</artifactId>
     </dependency>
-    <dependency>
-      <groupId>org.commonmark</groupId>
-      <artifactId>commonmark</artifactId>
-      <version>0.28.0</version>
-    </dependency>
 
     <!-- Compile-time dependencies -->
     <dependency>
@@ -66,6 +61,16 @@
       <artifactId>auto-service-annotations</artifactId>
       <optional>true</optional>
     </dependency>
+    <dependency>
+      <groupId>org.commonmark</groupId>
+      <artifactId>commonmark</artifactId>
+      <version>0.28.0</version>
+    </dependency>
+    <dependency>
+      <groupId>org.commonmark</groupId>
+      <artifactId>commonmark-ext-gfm-tables</artifactId>
+      <version>0.28.0</version>
+    </dependency>
 
     <!-- Test dependencies -->
     <dependency>
diff --git a/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocFormatter.java b/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocFormatter.java
index dbc2be841..f801a4d75 100644
--- a/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocFormatter.java
+++ b/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocFormatter.java
@@ -44,6 +44,7 @@
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownCodeSpanStart;
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownFencedCodeBlock;
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownHardLineBreak;
+import com.google.googlejavaformat.java.javadoc.Token.MarkdownTable;
 import com.google.googlejavaformat.java.javadoc.Token.MoeBeginStripComment;
 import com.google.googlejavaformat.java.javadoc.Token.MoeEndStripComment;
 import com.google.googlejavaformat.java.javadoc.Token.OptionalLineBreak;
@@ -137,6 +138,7 @@ private static String render(List<Token> input, int blockIndent, boolean classic
         case MarkdownHardLineBreak unused -> output.writeMarkdownHardLineBreak();
         case Literal t -> output.writeLiteral(t);
         case MarkdownFencedCodeBlock t -> output.writeMarkdownFencedCodeBlock(t);
+        case MarkdownTable t -> output.writeMarkdownTable(t);
         case ListItemCloseTag unused -> {}
         case OptionalLineBreak unused -> {}
         case ParagraphCloseTag unused -> {}
diff --git a/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocWriter.java b/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocWriter.java
index 182762a7f..53f6f1a29 100644
--- a/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocWriter.java
+++ b/core/src/main/java/com/google/googlejavaformat/java/javadoc/JavadocWriter.java
@@ -35,6 +35,7 @@
 import com.google.googlejavaformat.java.javadoc.Token.ListOpenTag;
 import com.google.googlejavaformat.java.javadoc.Token.Literal;
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownFencedCodeBlock;
+import com.google.googlejavaformat.java.javadoc.Token.MarkdownTable;
 import com.google.googlejavaformat.java.javadoc.Token.MoeBeginStripComment;
 import com.google.googlejavaformat.java.javadoc.Token.MoeEndStripComment;
 import com.google.googlejavaformat.java.javadoc.Token.PreCloseTag;
@@ -44,6 +45,7 @@
 import com.google.googlejavaformat.java.javadoc.Token.StartOfLineToken;
 import com.google.googlejavaformat.java.javadoc.Token.TableCloseTag;
 import com.google.googlejavaformat.java.javadoc.Token.TableOpenTag;
+import java.util.List;
 
 /**
  * Stateful object that accepts "requests" and "writes," producing formatted Javadoc.
@@ -354,6 +356,20 @@ void writeMarkdownFencedCodeBlock(MarkdownFencedCodeBlock token) {
     requestBlankLine();
   }
 
+  void writeMarkdownTable(MarkdownTable token) {
+    if (wroteAnythingSignificant && !atStartOfLine) {
+      requestBlankLine();
+    }
+    flushWhitespace();
+    List<String> lines = token.value().lines().toList();
+    output.append(lines.get(0));
+    for (String line : lines.subList(1, lines.size())) {
+      writeNewline(AutoIndent.NO_AUTO_INDENT);
+      output.append(line);
+    }
+    requestBlankLine();
+  }
+
   @Override
   public String toString() {
     return output.toString();
diff --git a/core/src/main/java/com/google/googlejavaformat/java/javadoc/MarkdownPositions.java b/core/src/main/java/com/google/googlejavaformat/java/javadoc/MarkdownPositions.java
index d5fdde5d9..0bbb56fee 100644
--- a/core/src/main/java/com/google/googlejavaformat/java/javadoc/MarkdownPositions.java
+++ b/core/src/main/java/com/google/googlejavaformat/java/javadoc/MarkdownPositions.java
@@ -28,11 +28,14 @@
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownCodeSpanEnd;
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownCodeSpanStart;
 import com.google.googlejavaformat.java.javadoc.Token.MarkdownFencedCodeBlock;
+import com.google.googlejavaformat.java.javadoc.Token.MarkdownTable;
 import com.google.googlejavaformat.java.javadoc.Token.ParagraphCloseTag;
 import com.google.googlejavaformat.java.javadoc.Token.ParagraphOpenTag;
 import java.util.Objects;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
+import org.commonmark.ext.gfm.tables.TableBlock;
+import org.commonmark.ext.gfm.tables.TablesExtension;
 import org.commonmark.node.BulletList;
 import org.commonmark.node.Code;
 import org.commonmark.node.FencedCodeBlock;
@@ -93,6 +96,10 @@ void visit(Node node) {
         case OrderedList orderedList -> addSpan(orderedList, LIST_OPEN_TOKEN, LIST_CLOSE_TOKEN);
         case ListItem listItem -> alreadyVisitedChildren = visitListItem(listItem);
         case FencedCodeBlock fencedCodeBlock -> visitFencedCodeBlock(fencedCodeBlock);
+        case TableBlock tableBlock -> {
+          visitTableBlock(tableBlock);
+          alreadyVisitedChildren = true;
+        }
         case Code code -> visitCodeSpan(code);
         // TODO: others
         default -> {}
@@ -143,6 +150,12 @@ private void visitFencedCodeBlock(FencedCodeBlock fencedCodeBlock) {
       positionToToken.get(start).addLast(token);
     }
 
+    private void visitTableBlock(TableBlock tableBlock) {
+      int start = startPosition(tableBlock);
+      int end = endPosition(tableBlock);
+      positionToToken.get(start).addLast(new MarkdownTable(input.substring(start, end)));
+    }
+
     private void visitCodeSpan(Code code) {
       int start = startPosition(code);
       int end = endPosition(code);
@@ -200,7 +213,10 @@ public String toString() {
   }
 
   private static final Parser PARSER =
-      Parser.builder().includeSourceSpans(IncludeSourceSpans.BLOCKS_AND_INLINES).build();
+      Parser.builder()
+          .includeSourceSpans(IncludeSourceSpans.BLOCKS_AND_INLINES)
+          .extensions(ImmutableList.of(TablesExtension.create()))
+          .build();
 
   private static final HeaderOpenTag HEADER_OPEN_TOKEN = new HeaderOpenTag("");
   private static final HeaderCloseTag HEADER_CLOSE_TOKEN = new HeaderCloseTag("");
diff --git a/core/src/main/java/com/google/googlejavaformat/java/javadoc/Token.java b/core/src/main/java/com/google/googlejavaformat/java/javadoc/Token.java
index 42b73b069..caa35d3fa 100644
--- a/core/src/main/java/com/google/googlejavaformat/java/javadoc/Token.java
+++ b/core/src/main/java/com/google/googlejavaformat/java/javadoc/Token.java
@@ -140,6 +140,20 @@ record MarkdownCodeSpanEnd(String value) implements Token {}
   record MarkdownFencedCodeBlock(String value, String start, String end, String literal)
       implements Token {}
 
+  /**
+   * A Markdown table, like:
+   *
+   * {@snippet :
+   * | foo | bar |
+   * | --- | --- |
+   * | baz | qux |
+   * }
+   *
+   * @param value the full text of the table as it appeared in the input, including the delimiters
+   *     and the literal content.
+   */
+  record MarkdownTable(String value) implements Token {}
+
   /**
    * Whitespace that is not in a {@code <pre>} or {@code <table>} section. Whitespace includes
    * leading newlines, asterisks, and tabs and spaces. In the output, it is translated to newlines
diff --git a/core/src/test/java/com/google/googlejavaformat/java/JavadocFormattingTest.java b/core/src/test/java/com/google/googlejavaformat/java/JavadocFormattingTest.java
index 02af3d137..5f2f12526 100644
--- a/core/src/test/java/com/google/googlejavaformat/java/JavadocFormattingTest.java
+++ b/core/src/test/java/com/google/googlejavaformat/java/JavadocFormattingTest.java
@@ -1516,15 +1516,6 @@ class Test {}
     doFormatTest(input, expected);
   }
 
-  private void doFormatTest(String input, String expected) {
-    try {
-      String actual = formatter.formatSource(input);
-      assertThat(actual).isEqualTo(expected);
-    } catch (FormatterException e) {
-      throw new AssertionError(e);
-    }
-  }
-
   @Test
   public void windowsLineSeparator() throws FormatterException {
     String input =
@@ -2016,6 +2007,8 @@ public void markdownTables() {
     assume().that(MARKDOWN_JAVADOC_SUPPORTED).isTrue();
     String input =
 """
+/// Table McTableface
+///
 /// | foo | bar |
 /// | --- | --- |
 /// | baz | qux |
@@ -2023,18 +2016,28 @@ public void markdownTables() {
 /// - |foo|bar|
 ///   |--:|:--|
 ///   |baz|qux|
+///
+/// - Another list.
+///
+///   | which | contains |
+///   | ----- | -------- |
+///   | a | table |
 class Test {}
 """;
-    // TODO: unmangle the tables
-    String expected =
-"""
-/// | foo | bar | | --- | --- | | baz | qux |
-/// - |foo|bar| |--:|:--| |baz|qux|
-class Test {}
-""";
+    // We don't currently try to align the column markers in the rows of the last table.
+    String expected = input;
     doFormatTest(input, expected);
   }
 
+  private void doFormatTest(String input, String expected) {
+    try {
+      String actual = formatter.formatSource(input);
+      assertThat(actual).isEqualTo(expected);
+    } catch (FormatterException e) {
+      throw new AssertionError(e);
+    }
+  }
+
   // TODO: b/346668798 - Test the following Markdown constructs, and make the tests work as needed.
   // We can assume that the CommonMark parser correctly handles Markdown, so the question is whether
   // they are subsequently mishandled by our formatting logic. So for example the CommonMark parser
@@ -2074,10 +2077,4 @@ class Test {}
   //
   // - Autolinks
   //   <http://example.com> should be preserved. https://spec.commonmark.org/0.31.2/#autolink
-  //
-  // - Tables
-  //   | foo | bar |
-  //   | --- | --- |
-  //   | baz | qux |
-  //   Probably we should just try not to mangle them. https://spec.commonmark.org/0.31.2/#tables
 }