diff --git a/docs/README.md b/docs/README.md
index 51dff51aec..f8b0eb0410 100644
--- a/docs/README.md
+++ b/docs/README.md
@@ -2068,7 +2068,7 @@ You can use the `--stream, -S` flag to make two things happen:
 1. The output is flushed in much smaller chunks without any buffering, which makes HTTPie behave kind of like `tail -f` for URLs.
 2. Streaming becomes enabled even when the output is prettified: It will be applied to each line of the response and flushed immediately. This makes it possible to have a nice output for long-lived requests, such as one to the [Twitter streaming API](https://developer.twitter.com/en/docs/tutorials/consuming-streaming-data).
 
-The `--stream` option is automatically enabled when the response headers include `Content-Type: text/event-stream`.
+The `--stream` option is automatically enabled when the response `Content-Type` is `text/event-stream`, `application/x-ndjson`, or `application/ndjson`.
 
 ### Example use cases
 
diff --git a/httpie/output/streams.py b/httpie/output/streams.py
index 811093808a..dd4640d686 100644
--- a/httpie/output/streams.py
+++ b/httpie/output/streams.py
@@ -216,12 +216,20 @@ def iter_body(self) -> Iterable[bytes]:
             yield self.process_body(line) + lf
             first_chunk = False
 
+    SSE_DATA_PREFIX = 'data:'
+    SSE_MIME = 'text/event-stream'
+    JSON_MIME = 'application/json'
+
     def process_body(self, chunk: Union[str, bytes]) -> bytes:
         if not isinstance(chunk, str):
             # Text when a converter has been used,
             # otherwise it will always be bytes.
             chunk = self.decode_chunk(chunk)
-        chunk = self.formatting.format_body(content=chunk, mime=self.mime)
+        mime = self.mime
+        if mime == self.SSE_MIME and chunk.lstrip().startswith(self.SSE_DATA_PREFIX):
+            # SSE data lines may contain JSON; use JSON mime for highlighting
+            mime = self.JSON_MIME
+        chunk = self.formatting.format_body(content=chunk, mime=mime)
         return smart_encode(chunk, self.output_encoding)
 
 
diff --git a/httpie/output/writer.py b/httpie/output/writer.py
index 4a2949bce2..23345b92e6 100644
--- a/httpie/output/writer.py
+++ b/httpie/output/writer.py
@@ -23,6 +23,12 @@
 MESSAGE_SEPARATOR = '\n\n'
 MESSAGE_SEPARATOR_BYTES = MESSAGE_SEPARATOR.encode()
 
+AUTO_STREAMING_CONTENT_TYPES = frozenset({
+    'text/event-stream',
+    'application/x-ndjson',
+    'application/ndjson',
+})
+
 
 def write_message(
     requests_message: RequestsMessage,
@@ -167,7 +173,7 @@ def get_stream_type_and_kwargs(
         raw_content_type_header = headers.get('Content-Type', None)
         if raw_content_type_header:
             content_type_header, _ = parse_content_type_header(raw_content_type_header)
-            is_stream = (content_type_header == 'text/event-stream')
+            is_stream = (content_type_header in AUTO_STREAMING_CONTENT_TYPES)
 
     if not env.stdout_isatty and not prettify_groups:
         stream_class = RawStream
diff --git a/tests/test_stream.py b/tests/test_stream.py
index b0b9b8bde8..db7c41be59 100644
--- a/tests/test_stream.py
+++ b/tests/test_stream.py
@@ -115,10 +115,9 @@ def test_redirected_stream(httpbin):
     assert BIN_FILE_CONTENT in r
 
 
-# /drip endpoint produces 3 individual lines,
-# if we set text/event-stream HTTPie should stream
-# it by default. Otherwise, it will buffer and then
-# print.
+# /drip endpoint produces 3 individual lines.
+# Auto-streaming content types (SSE, NDJSON) should stream
+# line by line; other types should buffer into a single write.
 @pytest.mark.parametrize('extras, expected', [
     (
         ['Accept:text/event-stream'],
@@ -128,6 +127,18 @@ def test_redirected_stream(httpbin):
         ['Accept:text/event-stream; charset=utf-8'],
         3
     ),
+    (
+        ['Accept:application/x-ndjson'],
+        3
+    ),
+    (
+        ['Accept:application/ndjson'],
+        3
+    ),
+    (
+        ['Accept:application/x-ndjson; charset=utf-8'],
+        3
+    ),
     (
         ['Accept:text/plain'],
         1
@@ -144,6 +155,35 @@ def test_auto_streaming(http_server, extras, expected):
     ]) == expected
 
 
+def test_ndjson_auto_streaming(http_server):
+    """NDJSON responses should auto-stream line by line."""
+    env = MockEnvironment()
+    env.stdout.write = Mock()
+    http(http_server + '/ndjson', env=env)
+    json_writes = [
+        call_arg
+        for call_arg in env.stdout.write.call_args_list
+        if 'message' in call_arg[0][0]
+    ]
+    assert len(json_writes) == 3
+
+
+def test_ndjson_json_formatting(http_server):
+    """NDJSON lines should be pretty-printed as JSON."""
+    r = http('--pretty=format', http_server + '/ndjson')
+    assert '"message": "test"' in r
+    assert '"index": 0' in r
+
+
+def test_sse_json_highlighting(http_server):
+    """SSE data lines containing JSON should get syntax highlighting."""
+    r = http('--pretty=colors', http_server + '/sse-json')
+    # With colors enabled, JSON keys/values should have ANSI escape codes
+    # The raw 'data:' prefix and JSON content should both appear
+    assert 'data:' in r
+    assert 'message' in r
+
+
 def test_streaming_encoding_detection(http_server):
     r = http('--stream', http_server + '/stream/encoding/random')
     assert ASCII_FILE_CONTENT in r
diff --git a/tests/utils/http_server.py b/tests/utils/http_server.py
index 86cc069c57..c401c06fc7 100644
--- a/tests/utils/http_server.py
+++ b/tests/utils/http_server.py
@@ -42,22 +42,53 @@ def get_headers(handler):
     handler.end_headers()
 
 
-@TestHandler.handler('GET', '/drip')
-def chunked_drip(handler):
+def _send_chunked_response(handler, content_type, chunks):
+    """Send a chunked HTTP response with the given content type and body chunks."""
     handler.send_response(200)
-    accept = handler.headers.get('Accept')
-    if accept is not None:
-        handler.send_header('Content-Type', accept)
+    if content_type is not None:
+        handler.send_header('Content-Type', content_type)
     handler.send_header('Transfer-Encoding', 'chunked')
     handler.end_headers()
 
-    for _ in range(3):
-        body = 'test\n'
+    for body in chunks:
         handler.wfile.write(f'{len(body):X}\r\n{body}\r\n'.encode('utf-8'))
 
     handler.wfile.write('0\r\n\r\n'.encode('utf-8'))
 
 
+@TestHandler.handler('GET', '/drip')
+def chunked_drip(handler):
+    _send_chunked_response(
+        handler,
+        content_type=handler.headers.get('Accept'),
+        chunks=['test\n' for _ in range(3)],
+    )
+
+
+@TestHandler.handler('GET', '/ndjson')
+def ndjson_stream(handler):
+    _send_chunked_response(
+        handler,
+        content_type='application/x-ndjson',
+        chunks=[
+            json.dumps({"index": i, "message": "test"}) + '\n'
+            for i in range(3)
+        ],
+    )
+
+
+@TestHandler.handler('GET', '/sse-json')
+def sse_json_stream(handler):
+    _send_chunked_response(
+        handler,
+        content_type='text/event-stream',
+        chunks=[
+            f'data: {json.dumps({"index": i, "message": "test"})}\n\n'
+            for i in range(3)
+        ],
+    )
+
+
 @TestHandler.handler('GET', '/stream/encoding/random')
 def random_encoding(handler):
     from tests.fixtures import ASCII_FILE_CONTENT, FILE_CONTENT as UNICODE_FILE_CONTENT