Environment variables

In general, environment variables, like LANGFLOW_PORT or LANGFLOW_LOG_LEVEL, configure how Langflow runs. These are broad settings that apply to your entire Langflow deployment.

In contrast, global variables are user-defined values stored in Langflow's database for use in flows, such as OPENAI_API_KEY. Langflow can also source global variables from environment variables. For more information, see Langflow global variables.

Configure environment variables for Langflow OSS

Langflow recognizes supported environment variables from the following sources:

• Environment variables set in your terminal.
• Environment variables imported from a .env file when starting Langflow.
• Environment variables set with the Langflow CLI, including the --env-file option and direct options, such as --port.

You can choose to use one or more of these sources.

Precedence

If the same environment variable is set in multiple places, the following hierarchy applies:

1. Langflow CLI options override all other sources.
2. The .env file overrides system environment variables.
3. System environment variables are used only if not set elsewhere. When running a Langflow Docker image, the -e flag can be used to set additional system environment variables.

For example:

• If you set LANGFLOW_PORT=8080 in your system environment and LANGFLOW_PORT=7860 in .env, Langflow uses 7860 from .env.
• If you use the Langflow CLI to run langflow run --env-file .env --port 9000, and you set LANGFLOW_PORT=7860 in .env, then Langflow uses 9000 from the CLI option.

Set environment variables in your terminal

Run the following commands to set environment variables for your current terminal session:

Linux or macOS

export VARIABLE_NAME='VALUE'

Windows

set VARIABLE_NAME='VALUE'

Docker

docker run -it --rm \
    -p 7860:7860 \
    -e VARIABLE_NAME='VALUE' \
    langflowai/langflow:latest

When you start Langflow, it looks for environment variables that you've set in your terminal. If it detects a supported environment variable, then it automatically adopts the specified value, subject to precedence rules.

Import environment variables from a .env file

1. If Langflow is running, quit Langflow.

2. Create a .env file, and then open it in your preferred editor.

3. Define Langflow environment variables in the .env file.

   Example: .env

   DO_NOT_TRACK=True
   LANGFLOW_AUTO_LOGIN=False
   LANGFLOW_AUTO_SAVING=True
   LANGFLOW_AUTO_SAVING_INTERVAL=1000
   LANGFLOW_BACKEND_ONLY=False
   LANGFLOW_BUNDLE_URLS=["https://github.com/user/repo/commit/hash"]
   LANGFLOW_CACHE_TYPE=async
   LANGFLOW_COMPONENTS_PATH=/path/to/components/
   LANGFLOW_COMPONENTS_INDEX_PATH=/path/to/component_index.json
   LANGFLOW_CONFIG_DIR=/path/to/config/
   LANGFLOW_DATABASE_URL=postgresql://user:password@localhost:5432/langflow
   LANGFLOW_DEV=False
   LANGFLOW_FALLBACK_TO_ENV_VAR=False
   LANGFLOW_HEALTH_CHECK_MAX_RETRIES=5
   LANGFLOW_HOST=localhost
   LANGFLOW_LANGCHAIN_CACHE=InMemoryCache
   LANGFLOW_MAX_FILE_SIZE_UPLOAD=10000
   LANGFLOW_MAX_ITEMS_LENGTH=100
   LANGFLOW_MAX_TEXT_LENGTH=1000
   LANGFLOW_LOG_LEVEL=error
   LANGFLOW_OPEN_BROWSER=False
   LANGFLOW_PORT=7860
   LANGFLOW_REMOVE_API_KEYS=False
   LANGFLOW_SAVE_DB_IN_CONFIG_DIR=True
   LANGFLOW_SECRET_KEY=somesecretkey
   LANGFLOW_STORE_ENVIRONMENT_VARIABLES=True
   LANGFLOW_SUPERUSER=adminuser
   LANGFLOW_SUPERUSER_PASSWORD=adminpass
   LANGFLOW_WORKER_TIMEOUT=300
   LANGFLOW_WORKERS=3

   For additional examples, see .env.example in the Langflow repository.

4. Save and close .env.

5. Start Langflow with your .env file:

   Local

   uv run langflow run --env-file .env

   Docker

   docker run -it --rm \
       -p 7860:7860 \
       --env-file .env \
       langflowai/langflow:latest

   If your .env file isn't in the same directory, provide the path to your .env file.

On startup, Langflow imports the environment variables from your .env file, as well as any others that you set in your terminal, and then adopts their specified values.

Configure environment variables for development

The following examples show how to configure Langflow using environment variables in different development scenarios.

.env file

The .env file is a text file that contains key-value pairs of environment variables.

Create or edit a .env file in the root directory of your application or Langflow environment, and then add your configuration variables to the file:

Example: .env

.env

DO_NOT_TRACK=True
LANGFLOW_AUTO_LOGIN=False
LANGFLOW_AUTO_SAVING=True
LANGFLOW_AUTO_SAVING_INTERVAL=1000
LANGFLOW_BACKEND_ONLY=False
LANGFLOW_BUNDLE_URLS=["https://github.com/user/repo/commit/hash"]
LANGFLOW_CACHE_TYPE=async
LANGFLOW_COMPONENTS_PATH=/path/to/components/
LANGFLOW_COMPONENTS_INDEX_PATH=/path/to/component_index.json
LANGFLOW_CONFIG_DIR=/path/to/config/
LANGFLOW_DATABASE_URL=postgresql://user:password@localhost:5432/langflow
LANGFLOW_DEV=False
LANGFLOW_FALLBACK_TO_ENV_VAR=False
LANGFLOW_HEALTH_CHECK_MAX_RETRIES=5
LANGFLOW_HOST=localhost
LANGFLOW_LANGCHAIN_CACHE=InMemoryCache
LANGFLOW_MAX_FILE_SIZE_UPLOAD=10000
LANGFLOW_MAX_ITEMS_LENGTH=100
LANGFLOW_MAX_TEXT_LENGTH=1000
LANGFLOW_LOG_LEVEL=error
LANGFLOW_OPEN_BROWSER=False
LANGFLOW_PORT=7860
LANGFLOW_REMOVE_API_KEYS=False
LANGFLOW_SAVE_DB_IN_CONFIG_DIR=True
LANGFLOW_SECRET_KEY=somesecretkey
LANGFLOW_STORE_ENVIRONMENT_VARIABLES=True
LANGFLOW_SUPERUSER=adminuser
LANGFLOW_SUPERUSER_PASSWORD=adminpass
LANGFLOW_WORKER_TIMEOUT=300
LANGFLOW_WORKERS=3

Systemd service

A systemd service configuration file configures Linux system services.

To add environment variables, create or edit a service configuration file and add an override.conf file. This file allows you to override the default environment variables for the service.

Example: override.conf

override.conf

[Service]
Environment="DO_NOT_TRACK=true"
Environment="LANGFLOW_AUTO_LOGIN=false"
Environment="LANGFLOW_AUTO_SAVING=true"
Environment="LANGFLOW_AUTO_SAVING_INTERVAL=1000"
Environment="LANGFLOW_BACKEND_ONLY=false"
Environment="LANGFLOW_BUNDLE_URLS=[\"https://github.com/user/repo/commit/hash\"]"
Environment="LANGFLOW_CACHE_TYPE=async"
Environment="LANGFLOW_COMPONENTS_PATH=/path/to/components/"
Environment="LANGFLOW_COMPONENTS_INDEX_PATH=/path/to/component_index.json"
Environment="LANGFLOW_CONFIG_DIR=/path/to/config"
Environment="LANGFLOW_DATABASE_URL=postgresql://user:password@localhost:5432/langflow"
Environment="LANGFLOW_DEV=false"
Environment="LANGFLOW_FALLBACK_TO_ENV_VAR=false"
Environment="LANGFLOW_HEALTH_CHECK_MAX_RETRIES=5"
Environment="LANGFLOW_HOST=localhost"
Environment="LANGFLOW_LANGCHAIN_CACHE=InMemoryCache"
Environment="LANGFLOW_MAX_FILE_SIZE_UPLOAD=10000"
Environment="LANGFLOW_MAX_ITEMS_LENGTH=100"
Environment="LANGFLOW_MAX_TEXT_LENGTH=1000"
Environment="LANGFLOW_LOG_ENV=container_json"
Environment="LANGFLOW_LOG_FILE=logs/langflow.log"
Environment="LANGFLOW_LOG_LEVEL=error"
Environment="LANGFLOW_OPEN_BROWSER=false"
Environment="LANGFLOW_PORT=7860"
Environment="LANGFLOW_REMOVE_API_KEYS=false"
Environment="LANGFLOW_SAVE_DB_IN_CONFIG_DIR=true"
Environment="LANGFLOW_SECRET_KEY=somesecretkey"
Environment="LANGFLOW_STORE_ENVIRONMENT_VARIABLES=true"
Environment="LANGFLOW_SUPERUSER=adminuser"
Environment="LANGFLOW_SUPERUSER_PASSWORD=adminpass"
Environment="LANGFLOW_WORKER_TIMEOUT=300"
Environment="LANGFLOW_WORKERS=3"

For more information on systemd, see the Red Hat documentation.

VSCode tasks.json

The tasks.json file located in .vscode/tasks.json is a configuration file for development environments using Visual Studio Code.

Create or edit the .vscode/tasks.json file in your project root.

Example: .vscode/tasks.json

.vscode/tasks.json

{
    "version": "2.0.0",
    "options": {
        "env": {
            "DO_NOT_TRACK": "true",
            "LANGFLOW_AUTO_LOGIN": "false",
            "LANGFLOW_AUTO_SAVING": "true",
            "LANGFLOW_AUTO_SAVING_INTERVAL": "1000",
            "LANGFLOW_BACKEND_ONLY": "false",
            "LANGFLOW_BUNDLE_URLS": "[\"https://github.com/user/repo/commit/hash\"]",
            "LANGFLOW_CACHE_TYPE": "async",
            "LANGFLOW_COMPONENTS_PATH": "D:/path/to/components/",
            "LANGFLOW_COMPONENTS_INDEX_PATH": "D:/path/to/component_index.json",
            "LANGFLOW_CONFIG_DIR": "D:/path/to/config/",
            "LANGFLOW_DATABASE_URL": "postgresql://postgres:password@localhost:5432/langflow",
            "LANGFLOW_DEV": "false",
            "LANGFLOW_FALLBACK_TO_ENV_VAR": "false",
            "LANGFLOW_HEALTH_CHECK_MAX_RETRIES": "5",
            "LANGFLOW_HOST": "localhost",
            "LANGFLOW_LANGCHAIN_CACHE": "InMemoryCache",
            "LANGFLOW_MAX_FILE_SIZE_UPLOAD": "10000",
            "LANGFLOW_MAX_ITEMS_LENGTH": "100",
            "LANGFLOW_MAX_TEXT_LENGTH": "1000",
            "LANGFLOW_LOG_ENV": "container_csv",
            "LANGFLOW_LOG_FILE": "langflow.log",
            "LANGFLOW_LOG_LEVEL": "error",
            "LANGFLOW_OPEN_BROWSER": "false",
            "LANGFLOW_PORT": "7860",
            "LANGFLOW_REMOVE_API_KEYS": "true",
            "LANGFLOW_SAVE_DB_IN_CONFIG_DIR": "false",
            "LANGFLOW_SECRET_KEY": "somesecretkey",
            "LANGFLOW_STORE_ENVIRONMENT_VARIABLES": "true",
            "LANGFLOW_SUPERUSER": "adminuser",
            "LANGFLOW_SUPERUSER_PASSWORD": "adminpass",
            "LANGFLOW_WORKER_TIMEOUT": "300",
            "LANGFLOW_WORKERS": "3"
        }
    },
    "tasks": [
        {
            "label": "langflow backend",
            "type": "shell",
            "command": ". ./langflownightly/Scripts/activate && langflow run",
            "isBackground": true,
            "problemMatcher": []
        }
    ]
}

To run Langflow using the above VSCode tasks.json file, in the VSCode command palette, select Tasks: Run Task > langflow backend.

Set environment variables for Langflow Desktop

Environment variables set in your terminal aren't automatically available to GUI-based applications like Langflow Desktop when you launch them from the Windows or macOS GUI.

To modify environment variables for Langflow Desktop, set environment variables in a Desktop .env file, and then restart the app.

macOS

To modify the macOS .env file, do the following:

1. Create or edit ~/.langflow/data/.env.

2. Add your Langflow environment variables, for example:

   LANGFLOW_LOG_LEVEL=info
   LANGFLOW_DOCLING=true

3. Save the file.

4. Restart Langflow Desktop.

Windows .env file

To modify the Windows .env file, do the following:

1. Create or edit %APPDATA%\com.LangflowDesktop\data\.env.

2. Add your Langflow environment variables, for example:

   LANGFLOW_LOG_LEVEL=info
   LANGFLOW_DOCLING=true

3. Save the file.

4. Restart Langflow Desktop.

Windows user environment variables

Windows supports two sources for Langflow Desktop environment variables: a Langflow application .env file, and Windows user environment variables.

The .env file at %APPDATA%\com.LangflowDesktop\data\.env is the recommended approach, but Windows user variables are useful for single-sourcing API keys between Langflow and other Windows applications. If the same variable is defined in both the Langflow application .env file and as a Windows user environment variable, the .env file takes precedence.

To modify the Windows user environment variables, do the following:

1. Press Win + R, enter sysdm.cpl, and then press Enter.
2. Click the Advanced tab, and then click Environment Variables.
3. In User variables, click New.
4. Enter the variable name, such as OPENAI_API_KEY, and its value.
5. Click OK, and then restart Langflow Desktop.

Supported environment variables

The following sections provide information about specific Langflow environment variables.

Authentication and security

See API keys and authentication.

Global variables

For information about the relationship between Langflow global variables and environment variables, as well as environment variables that control handling of global variables, see Global variables.

Logs

See Configure log options.

MCP servers

See Use Langflow as an MCP server.

A2A agents

See A2A environment variables.

Multi-worker deployments

For multi-worker tuning—including the Redis job queue and Gunicorn worker settings—see Deploy Langflow with multiple workers.

Monitoring and metrics

For OpenTelemetry traces, metrics, and logs that describe Langflow service health, including LANGFLOW_OTEL_* and OTEL_EXPORTER_OTLP_*, see OpenTelemetry.

For prompt-level LLM tracing, see Langfuse and the other monitoring integrations. For Prometheus on Kubernetes, see Best practices for Langflow on Kubernetes.

Server

The following environment variables set base Langflow server configuration, such as where the server is hosted, required files for SSL encryption, and the deployment type (frontend and backend, backend-only, development mode).

Variable	Format	Default	Description
LANGFLOW_​HOST	String	localhost	The host on which the Langflow server will run.
LANGFLOW_​PORT	Integer	7860	The port on which the Langflow server runs. The server automatically selects a free port if the specified port is in use.
LANGFLOW_​BACKEND_​ONLY	Boolean	False	Run only the Langflow backend service (no frontend).
LANGFLOW_​DEV	Boolean	False	Whether to run Langflow in development mode (may contain bugs).
LANGFLOW_​DEPLOYMENT_​PROFILE	String	dev	Deployment profile. dev skips production checks. prod runs production preflight before the server starts and aborts boot if a required service is missing.
LANGFLOW_​OPEN_​BROWSER	Boolean	False	Open the system web browser on startup.
LANGFLOW_​HEALTH_​CHECK_​MAX_​RETRIES	Integer	5	Set the maximum number of retries for Langflow's server status health checks.
LANGFLOW_​WORKERS	Integer	1	Number of worker processes. See Deploy Langflow with multiple workers.
LANGFLOW_​WORKER_​TIMEOUT	Integer	300	Worker timeout in seconds. See Deploy Langflow with multiple workers.
LANGFLOW_​GUNICORN_​PRELOAD	Boolean	False	Enables Gunicorn preload_​app. Non-Windows only. See Deploy Langflow with multiple workers.
LANGFLOW_​JOB_​QUEUE_​TYPE	String	asyncio	Job queue backend. Use redis for multi-worker deployments. See Deploy Langflow with multiple workers.
LANGFLOW_​DANGEROUSLY_​ALLOW_​MULTI_​WORKER_​WITHOUT_​SHARED_​QUEUE	Boolean	False	Allow startup with LANGFLOW_​WORKERS greater than 1 while LANGFLOW_​JOB_​QUEUE_​TYPE is asyncio. Unsupported with the Langflow UI and with MCP over SSE. See Run multiple workers without Redis.
LANGFLOW_​SSL_​CERT_​FILE	String	Not set	Path to the SSL certificate file for enabling HTTPS on the Langflow web server. This is separate from database SSL connections.
LANGFLOW_​SSL_​KEY_​FILE	String	Not set	Path to the SSL key file for enabling HTTPS on the Langflow web server. This is separate from database SSL connections.
LANGFLOW_​DEACTIVATE_​TRACING	Boolean	False	Deactivate tracing functionality.
LANGFLOW_​CELERY_​ENABLED	Boolean	False	Enable Celery for distributed task processing.
LANGFLOW_​ALEMBIC_​LOG_​TO_​STDOUT	Boolean	False	Whether to log Alembic database migration output to stdout instead of a log file. If true, Alembic logs to stdout and the default log file is ignored.

Storage

For file storage environment variables, see File storage environment variables.

For database environment variables, including PostgreSQL configuration, see Memory management options.

Telemetry

Anonymous product usage telemetry is separate from OpenTelemetry export.

See Telemetry to opt out with DO_NOT_TRACK.

OpenTelemetry

Langflow can export service-health traces, metrics, and logs over OTLP to your backend. This is not LLM tracing and is not the anonymous product telemetry above.

Set OTEL_EXPORTER_OTLP_ENDPOINT to enable export. If it is unset, Langflow installs no providers.

Variable	Format	Default	Description
OTEL_​EXPORTER_​OTLP_​ENDPOINT	String	Not set	OTLP collector or APM endpoint. Required to enable export.
LANGFLOW_​OTEL_​DB_​SPANS	Boolean	True	Whether to export SQLAlchemy database spans. Set to false to send flow and request spans only.
LANGFLOW_​OTEL_​LOG_​LEVEL	String	INFO	Minimum severity exported over OTLP. DEBUG exports more records (flow inputs and outputs are logged at DEBUG) but withholds their bodies unless LANGFLOW_​OTEL_​LOG_​BODIES=all is also set.
LANGFLOW_​OTEL_​LOG_​BODIES	String	allowlist	Set to all to export log message bodies, including completions and chat history.

Standard OpenTelemetry SDK variables such as OTEL_SERVICE_NAME and OTEL_EXPORTER_OTLP_HEADERS also apply. To verify the pipeline, run langflow observability doctor.

For the full list, what Langflow does and does not export, and vendor setup, see OpenTelemetry.

Sandbox backends

These variables control optional execution backends for the Python Interpreter Component and the legacy Python REPL tool. The exec-sandbox backend is not included in a default uv pip install langflow installation; install langflow[sandbox] to use it.

Variable	Format	Default	Description
LANGFLOW_​SANDBOX_​BACKEND	String	none	Name of the execution backend for the code-execution components (Python Interpreter and the legacy Python REPL tool). Any registered backend name is accepted — see Sandbox backends. Set to exec-sandbox to run user code in a dedicated QEMU microVM per execution on this host. Requires uv pip install 'langflow[sandbox]', Python 3.12 or later, and QEMU 8+ with hardware virtualization (KVM on Linux, HVF on macOS). When a configured backend is unavailable, execution fails with an error instead of falling back to in-process execution.
LANGFLOW_​SANDBOX_​TIMEOUT_​SECONDS	Integer	30	Wall-clock limit for one sandboxed execution (1-300 seconds). The selected backend can declare a lower cap, and a value above that cap is refused even inside this range — see Sandbox backends. Only used when LANGFLOW_​SANDBOX_​BACKEND is not none.
LANGFLOW_​SANDBOX_​MEMORY_​MB	Integer	192	Guest VM memory for sandboxed executions, in MB (minimum 128). Only used when LANGFLOW_​SANDBOX_​BACKEND is not none.
LANGFLOW_​SANDBOX_​ALLOW_​NETWORK	Boolean	False	Whether sandboxed code may access the network. The default runs the microVM fully offline. When enabled without LANGFLOW_​SANDBOX_​ALLOWED_​DOMAINS, exec-sandbox reaches only its package-registry defaults (PyPI). Only used when LANGFLOW_​SANDBOX_​BACKEND is not none.
LANGFLOW_​SANDBOX_​ALLOWED_​DOMAINS	String	Not set	Comma-separated list of domains sandboxed code may reach when network access is enabled. Treat like an SSRF allow-list: prefer narrow, fully-qualified domains. Only used when LANGFLOW_​SANDBOX_​BACKEND is not none.
LANGFLOW_​SANDBOX_​ALLOW_​SOFTWARE_​EMULATION	Boolean	False	Permit the sandbox to run without a hardware hypervisor, letting QEMU fall back to TCG software emulation. Upstream documents TCG as not security-supported, so enable only for trusted/development workloads. Only used when LANGFLOW_​SANDBOX_​BACKEND is not none.
LANGFLOW_​SANDBOX_​BACKEND_​PLUGINS	String	Not set	Comma-separated names of out-of-tree sandbox backends that may load, from the lfx.sandbox_​backends entry point group — see Sandbox backends. Empty (default) means no third-party backend code is imported at all. Loading a plugin imports its code into the Langflow process on the path that decides whether user code is isolated, and a backend's declared capabilities are configuration metadata rather than proof it enforces them, so naming one here is a statement that you trust that package. A name already used by a built-in backend is refused rather than replaced.

Visual editor and Playground behavior

Variable	Format	Default	Description
LANGFLOW_​AUTO_​SAVING	Boolean	True	Whether to automatically save the current flow draft. Auto-save only updates the live draft, and doesn't create saved flow versions.
LANGFLOW_​AUTO_​SAVING_​INTERVAL	Integer	1000	Set the auto-save interval in milliseconds if LANGFLOW_​AUTO_​SAVING=True.
LANGFLOW_​BUNDLE_​URLS	List[String]	[]	A list of URLs from which to load custom bundles. Supports GitHub URLs. If LANGFLOW_​AUTO_​LOGIN=True, flows from these bundles are loaded into the database.
LANGFLOW_​COMPONENTS_​PATH	String	Not set	Path to a directory containing custom components. Typically used if you have local custom components or you are building a Docker image with custom components.
LANGFLOW_​COMPONENTS_​INDEX_​PATH	String	Not set	File path or URL (http:// or https://) to a prebuilt component index JSON file used to populate built-in components in the visual editor. When not set, Langflow uses the included index. Useful for supplying a curated component index, for example in airgapped deployments. For more information, see Block custom components.
LANGFLOW_​AGENTIC_​EXPERIENCE	Boolean	True	Whether Langflow Assistant is available. Set to false to turn it off for a deployment, for example if you don't want model-generated component code running on your server. The Assistant's endpoints then return 404 and its run_​assistant MCP tool is unavailable, but the rest of the MCP toolkit at /api/v1/agentic/mcp is unaffected. To block code execution itself, including hand-written custom components, use LANGFLOW_​ALLOW_​CUSTOM_​COMPONENTS.
LANGFLOW_​ASSISTANT_​MAX_​MESSAGE_​LENGTH	Integer	2000	Maximum length, in characters, of a single message sent to Langflow Assistant. The limit is enforced by the Assistant API and mirrored to the visual editor, which blocks sending a longer message instead of truncating it. Raise it for users who paste long specifications into the Assistant; longer messages cost more per turn.
LANGFLOW_​ASSISTANT_​HISTORY_​TURNS	Integer	6	How many prior conversation turns Langflow Assistant injects into its prompt. The block is re-sent on every agent iteration, so this is a direct per-request token-cost lever: raise it for longer memory and accept the cost, or set 0 to inject no history. Values that aren't a number fall back to the default.
LANGFLOW_​ASSISTANT_​ITERATIONS	Integer	30	The step budget (max_​iterations) each Langflow Assistant agent runs with, which also derives LangGraph's recursion limit (max_​iterations * 2 + 5). Raise it when your flows are multi-stage by nature and the Assistant keeps reporting that it ran out of steps; a larger budget raises the worst-case token spend per turn. A user's /iterations N command overrides it for their session, and values are clamped to 1-200. Values that aren't a number fall back to the default.
LANGFLOW_​ASSISTANT_​VERIFY_​FLOWS	Boolean	True	Whether Langflow Assistant runs a flow it just built before presenting it, repairing fixable failures and adding a caveat to the reply when it can't. Set to false to deliver flows unverified, which is faster and cheaper but removes the guarantee that a delivered flow ran at least once.
LANGFLOW_​ALLOW_​CUSTOM_​COMPONENTS	Boolean	True	If false, disables custom components, in-editor editing of component code, and registered built-in code-execution components at runtime. This feature is in beta. For more information, see Block custom components and Component hardening for untrusted users.
LANGFLOW_​ALLOW_​PUBLIC_​CUSTOM_​COMPONENTS	Boolean	False	Whether flows shared through the Shareable Playground can run custom component code. By default, an unauthenticated public flow build runs only the server's own copy of each component, and a flow containing a custom component is refused with This flow cannot be executed. Set to true to build public flows from the database as their owner. This requires LANGFLOW_​ALLOW_​CUSTOM_​COMPONENTS=true, and it lets every node in a public flow run the code stored in that flow, not only the custom ones. To run a custom component in a shared Playground without relaxing that, install the component on the server with LANGFLOW_​COMPONENTS_​PATH instead. For more information, see Custom components in publicly shared flows.
LANGFLOW_​ALLOW_​COMPONENTS_​PATHS_​OVERRIDE	Boolean	True	When false alongside LANGFLOW_​ALLOW_​CUSTOM_​COMPONENTS=false, components contributed by LANGFLOW_​COMPONENTS_​PATH and LANGFLOW_​COMPONENTS_​INDEX_​PATH no longer bypass the block. Has no effect when LANGFLOW_​ALLOW_​CUSTOM_​COMPONENTS=true. For more information, see Block custom components.
LANGFLOW_​SUBSTITUTE_​OUTDATED_​COMPONENT_​CODE	Boolean	True	Only applies when LANGFLOW_​ALLOW_​CUSTOM_​COMPONENTS=false, where a flow's stored component code never runs anyway. By default, a node whose component type this server knows but whose stored code has drifted across an upgrade is rebuilt with this server's copy of that component, and the substitution is logged. Set to false to refuse those flows until each component is updated. Unrecognized component types are refused either way. For more information, see Block custom components.
LANGFLOW_​BLOCK_​CODE_​INTERPRETER_​COMPONENTS	Boolean	False	If true, blocks built-in code-execution components (Python Interpreter, REPL/Code tools, Smart Transform, and code-running agents). See Component hardening for untrusted users.
LANGFLOW_​RESTRICT_​LOCAL_​FILE_​ACCESS	Boolean	False	If true, confines built-in file components to the upload storage directory and blocks local-file SQL dialects and local Git clones. See Component hardening for untrusted users.
LANGFLOW_​MCP_​SERVER_​DOCKER_​HARDENING	Boolean	False	If true, applies a strict Docker-argument policy to MCP stdio servers that use the docker transport. See Component hardening for untrusted users.
LANGFLOW_​MCP_​SERVER_​ENV_​ALLOWLIST	String	Not set	Comma-separated allowlist of environment-variable names an MCP stdio config may set. Production preflight warns when this is unset. An empty value blocks all tenant-supplied environment variables. See MCP server environment variables.
LANGFLOW_​TWEAKS_​POLICY	String	permissive	Which component fields a run request may set through tweaks. For more information, see Restrict API tweaks.
LANGFLOW_​LOAD_​FLOWS_​PATH	String	Not set	Path to a directory containing flow JSON files to be loaded on startup. Typically used when creating a Docker image with prepackaged flows. Requires LANGFLOW_​AUTO_​LOGIN=True.
LANGFLOW_​LOAD_​FLOWS_​OVERWRITE_​ON_​NAME_​MATCH	Boolean	False	When a flow file in LANGFLOW_​LOAD_​FLOWS_​PATH shares a name with an existing DB row but has a different id, controls whether to overwrite the existing row. False (default) skips with a warning so UI edits are preserved on restart when file UUIDs regenerate. Set to True to opt into prepackaged-flows-are-source-of-truth semantics, typically for CI/CD pipelines.
LANGFLOW_​LOAD_​FLOWS_​PRESERVE_​VARIABLE_​BINDINGS	Boolean	True	Preserve global-variable bindings configured in the visual editor when an existing flow is reloaded from LANGFLOW_​LOAD_​FLOWS_​PATH. Bindings explicitly configured in the flow file still take precedence. Set to False to restore blind replacement of the flow data on restart.
LANGFLOW_​CREATE_​STARTER_​PROJECTS	Boolean	True	Whether to create templates during initialization. If false, Langflow doesn't create templates, and LANGFLOW_​UPDATE_​STARTER_​PROJECTS is treated as false.
LANGFLOW_​UPDATE_​STARTER_​PROJECTS	Boolean	True	Whether to update templates with the latest component versions when initializing after an upgrade.
LANGFLOW_​LAZY_​LOAD_​COMPONENTS	Boolean	False	If true, Langflow only partially loads components at startup and fully loads them on demand. This significantly reduces startup time but can cause a slight delay when a component is first used.
LANGFLOW_​EVENT_​DELIVERY	String	streaming	How to deliver build events to the frontend: polling, streaming or direct.
LANGFLOW_​FRONTEND_​PATH	String	./frontend	Path to the frontend directory containing build files. For development purposes only when you need to serve specific frontend code.
LANGFLOW_​FS_​TOOL_​BASE_​DIR	String	~/.langflow/fs_​tool/fs_​sandbox	Base directory for the File System component sandbox. All agent file operations are confined to this directory.
LANGFLOW_​MAX_​ITEMS_​LENGTH	Integer	100	Maximum number of items to store and display in the visual editor. Lists longer than this will be truncated when displayed in the visual editor. Doesn't affect outputs or data passed between components.
LANGFLOW_​MAX_​TEXT_​LENGTH	Integer	1000	Maximum number of characters to store and display in the visual editor. Responses longer than this will be truncated when displayed in the visual editor. Doesn't truncate outputs or responses passed between components.
LANGFLOW_​MAX_​TRANSACTIONS_​TO_​KEEP	Integer	3000	Maximum number of flow transaction events to keep in the database.
LANGFLOW_​MAX_​VERTEX_​BUILDS_​TO_​KEEP	Integer	3000	Maximum number of vertex builds to keep in the database. Relates to Playground functionality.
LANGFLOW_​MAX_​VERTEX_​BUILDS_​PER_​VERTEX	Integer	2	Maximum number of builds to keep per vertex. Older builds are deleted. Relates to Playground functionality.
LANGFLOW_​PUBLIC_​FLOW_​CLEANUP_​INTERVAL	Integer	3600	The interval in seconds at which data for shared Playground flows are cleaned up. Default: 3600 seconds (1 hour). Minimum: 600 seconds (10 minutes).
LANGFLOW_​PUBLIC_​FLOW_​EXPIRATION	Integer	86400	The time in seconds after which a shared Playground flow is considered expired and eligible for cleanup. Default: 86400 seconds (24 hours). Minimum: 600 seconds (10 minutes).

Hide UI elements with embedded mode

When you embed the Langflow visual editor in another application, such as an iframe within a host portal, you can hide UI elements that are usually visible in the visual editor.

The embedded mode environment variables control which UI elements appear in the visual editor. Embedded mode controls UI visibility only, and does not block API endpoints or deploy Langflow.

Embedded mode is distinct from the embedded chat widget, which embeds a single flow's chat UI in a website.

When LANGFLOW_EMBEDDED_MODE=true, Langflow hides the following buttons in the UI:

• Logout
• New project
• New flow
• Starter projects

You can hide individual elements without enabling the umbrella flag.

Variable	Format	Default	Description
LANGFLOW_​EMBEDDED_​MODE	Boolean	False	Umbrella flag for embedded or iframe deployments. When true, hides standalone UI elements: logout, new project, new flow, and starter projects. Individual hide flags can also be set without enabling this flag.
LANGFLOW_​HIDE_​LOGOUT_​BUTTON	Boolean	False	If true, hides the logout button in the account menu. Automatically enabled when LANGFLOW_​EMBEDDED_​MODE=true.
LANGFLOW_​HIDE_​NEW_​PROJECT_​BUTTON	Boolean	False	If true, hides the new project/folder button in the sidebar. Automatically enabled when LANGFLOW_​EMBEDDED_​MODE=true.
LANGFLOW_​HIDE_​NEW_​FLOW_​BUTTON	Boolean	False	If true, hides the new flow button in the header. Automatically enabled when LANGFLOW_​EMBEDDED_​MODE=true.
LANGFLOW_​HIDE_​STARTER_​PROJECTS	Boolean	False	If true, hides the starter projects tab in the templates modal. Does not affect database seeding of starter projects. Automatically enabled when LANGFLOW_​EMBEDDED_​MODE=true.
LANGFLOW_​HIDE_​GETTING_​STARTED_​PROGRESS	Boolean	False	If true, hides the getting-started onboarding progress UI. Not automatically enabled when LANGFLOW_​EMBEDDED_​MODE=true.