Langflow Module 6: Calling Flows from Code with the REST API

Step 1: Create a Langflow API key

Click your profile icon, choose Settings, then Langflow API Keys, and click Add New. Name the key and click Create API Key, then copy it immediately and store it securely. This key authenticates calls to Langflow and is separate from your model provider's key.

Step 2: Find the flow ID and a ready-made snippet

Open the flow, click Share, and choose API access. It shows ready-made snippets with your flow ID filled in. You can also give the flow a friendlier Endpoint Name there, using letters, numbers, hyphens, and underscores.

Step 3: Call the run endpoint with curl

Send a POST request to the flow's run endpoint with your key in the x-api-key header. The body carries the user's message in input_value. A session_id groups messages into one conversation.

export LANGFLOW_SERVER_URL="http://localhost:7860"
export FLOW_ID="your-flow-id"
export LANGFLOW_API_KEY="your-langflow-api-key"

curl -X POST "$LANGFLOW_SERVER_URL/api/v1/run/$FLOW_ID" \
  -H "Content-Type: application/json" \
  -H "x-api-key: $LANGFLOW_API_KEY" \
  -d '{"input_value": "Tell me about something interesting!",
       "session_id": "chat-123",
       "input_type": "chat", "output_type": "chat"}'

Step 4: Call it from Python

The same request works with the requests library. Read the keys from environment variables rather than writing them into source code. The reply text sits deep in the response structure, as shown in the Langflow quickstart.

import os
import requests

base = os.environ["LANGFLOW_SERVER_URL"]
flow_id = os.environ["FLOW_ID"]

response = requests.post(
    base + "/api/v1/run/" + flow_id,
    headers={
        "Content-Type": "application/json",
        "x-api-key": os.environ["LANGFLOW_API_KEY"],
    },
    json={
        "input_value": "Tell me about something interesting!",
        "session_id": "chat-123",
        "input_type": "chat",
        "output_type": "chat",
    },
    timeout=60,
)
response.raise_for_status()
data = response.json()
print(data["outputs"][0]["outputs"][0]["outputs"]["message"]["message"])

Step 5: Use sessions, streaming, and tweaks

Reuse the same session_id to continue a conversation. Add ?stream=true to the URL for streamed output. Tweaks override a component's parameters for a single run, using the component's ID as the key, which is useful for switching models or settings without editing the flow. Which tweaks are allowed is controlled by the LANGFLOW_TWEAKS_POLICY setting.

{
  "input_value": "Summarize this ticket",
  "tweaks": {
    "Agent-ZOknz": {"model_name": "your-model-name"}
  }
}