Langflow OSS 1.11 is released! This version adds Human-in-the-Loop for gated tool calls and reviews, A2A protocol support, AGUI-compatible streaming on the Workflow API, and more!
Before you upgrade to version 1.11, there are a few potentially breaking changes to be aware of.
The legacy langflow default password is removed, and components requiring the PyTorch dependency are not included by default. For more information, see the release notes.
Human-in-the-Loop
HITL pauses a flow, creates a stateful checkpoint, and waits for a human decision.
For more information, see Human-in-the-Loop.
A2A (Agent to Agent) protocol support
Langflow now supports the Agent2Agent (A2A) protocol. You can publish a flow so other agents can call it, and call remote A2A agents from inside a flow.
A2A support is off by default. To enable it, set the following environment variable before starting Langflow:
LANGFLOW_A2A_ENABLED=true
For more information, see Use Langflow as an A2A server.
AG-UI streaming for the Workflow API
POST /api/v2/workflows stream mode now supports stream_protocol: "agui" for AG-UI clients, in addition to the default Langflow EventManager protocol.
This example runs a flow in stream mode with the AG-UI protocol and prints each SSE event as it arrives:
import os
import requests
base = os.environ["LANGFLOW_SERVER_URL"]
headers = {
"Content-Type": "application/json",
"x-api-key": os.environ["LANGFLOW_API_KEY"],
}
payload = {
"flow_id": os.environ["FLOW_ID"],
"input_value": "Hello from an AG-UI client",
"mode": "stream",
"stream_protocol": "agui",
"session_id": "thread-123",
}
with requests.post(
f"{base}/api/v2/workflows",
headers=headers,
json=payload,
stream=True,
timeout=120,
) as response:
response.raise_for_status()
for line in response.iter_lines(decode_unicode=True):
if line:
print(line)
For more information, see Stream with AG-UI.
Unified Data Operations component
Text Operations, JSON Operations, and Table Operations are consolidated into a single Data Operations component. Saved flows that use the separate components continue to work.
New provider components
New NextPlaid, Paddle, Oracle, Valkey, and EmpirioLabs component bundles are available.
LFX can be installed as engine-only
uv pip install lfx installs the Langflow executor only, with no bundle components.
uv pip install langflow is unchanged.
For more information, see Extensions overview.
There's more to come
For the full list of changes in Langflow OSS 1.11, see the release notes or the full release on GitHub.
We're committed to continuously improving Langflow, and your input is crucial to this process. If you want to get more involved you can:
- Join our Discord community to network with other developers
- Explore our GitHub repository to contribute to the project or 🌟 the project
- Subscribe to our YouTube channel for step-by-step guides on using Langflow
- Stay updated on the latest AI trends and insights by subscribing to the AI++ newsletter
We're excited to see the projects you'll create with the Langflow 1.11 open-source package. Check out the 1.11.0 release on GitHub and get building!
Langflow Desktop 1.11 will be released soon. In the meantime, visit Download Langflow Desktop to install the current desktop version.
