Environment Variables

Environment Variable	Description	Default Value	Required	Example
Core Configuration
PUBLIC_ORIGIN	Public origin for generating complete URLs, must be set to your app’s access address	-	Yes	https://app.teable.ai
SECRET_KEY	Key for JWT, sessions, and sharing, use a strong password	defaultSecretKey	Yes	yourStrongSecretKey
BACKEND_STORAGE_ENCRYPTION_KEY	Encryption key for storage tokens. The default is a public constant in the source — production deployments must set their own random value (the full-featured assets generate one per host)	public constant	-	16-char random string
BACKEND_STORAGE_ENCRYPTION_IV	Encryption IV paired with BACKEND_STORAGE_ENCRYPTION_KEY	public constant	-	16-char random string
BACKEND_ACCESS_TOKEN_ENCRYPTION_KEY	Encryption key for personal access tokens. Same caveat: the default is public — set your own in production	public constant	-	16-char random string
BACKEND_ACCESS_TOKEN_ENCRYPTION_IV	Encryption IV paired with BACKEND_ACCESS_TOKEN_ENCRYPTION_KEY	public constant	-	16-char random string
PORT	Port on which the application runs	3000	-	3000
LOG_LEVEL	Log level, options: fatal, error, warn, info, debug, trace	info	-	debug
Storage Configuration
BACKEND_STORAGE_PROVIDER	Storage provider, options: local, minio, s3, aliyun	local	-	s3
BACKEND_STORAGE_LOCAL_PATH	Local storage path	.assets/uploads	-	.assets/uploads
BACKEND_STORAGE_PUBLIC_BUCKET	Public bucket name	public	-	teable-public
BACKEND_STORAGE_PRIVATE_BUCKET	Private bucket name for attachments, app files, record-history cold parts, and other protected content	private	-	teable-private
BACKEND_STORAGE_PUBLIC_URL	Public URL override for public bucket (optional)	-	-	https://cdn.example.com
BACKEND_STORAGE_PRIVATE_BUCKET_ENDPOINT	Private bucket endpoint override (optional)	-	-	https://private-bucket.endpoint
BACKEND_STORAGE_S3_REGION	S3 storage region, required when BACKEND_STORAGE_PROVIDER is s3	-	-	us-east-2
BACKEND_STORAGE_S3_ENDPOINT	S3 storage endpoint, required when BACKEND_STORAGE_PROVIDER is s3	-	-	https://s3.us-east-2.amazonaws.com
BACKEND_STORAGE_S3_INTERNAL_ENDPOINT	S3 internal endpoint (optional)	-	-	http://s3.internal
BACKEND_STORAGE_S3_ACCESS_KEY	S3 storage access key, required when BACKEND_STORAGE_PROVIDER is s3	-	-	your_access_key
BACKEND_STORAGE_S3_SECRET_KEY	S3 storage secret key, required when BACKEND_STORAGE_PROVIDER is s3	-	-	your_secret_key
BACKEND_STORAGE_S3_MAX_SOCKETS	S3 max sockets (optional)	100	-	100
BACKEND_STORAGE_MINIO_ENDPOINT	MinIO storage endpoint	-	-	minio.example.com
BACKEND_STORAGE_MINIO_PORT	MinIO port	9000	-	443
BACKEND_STORAGE_MINIO_USE_SSL	Whether MinIO uses SSL	false	-	true
BACKEND_STORAGE_MINIO_ACCESS_KEY	MinIO access key	-	-	access-key
BACKEND_STORAGE_MINIO_SECRET_KEY	MinIO secret key	-	-	secret-key
BACKEND_STORAGE_MINIO_INTERNAL_ENDPOINT	MinIO internal endpoint (optional, no HTTPS)	-	-	minio.internal
BACKEND_STORAGE_MINIO_INTERNAL_PORT	MinIO internal port (optional)	9000	-	9000
BACKEND_STORAGE_MINIO_REGION	MinIO region (optional)	-	-	us-east-1
BACKEND_STORAGE_UPLOAD_METHOD	Upload method	put	-	put
STORAGE_PREFIX	Storage prefix, defaults to PUBLIC_ORIGIN	PUBLIC_ORIGIN	-	http://localhost:3000
Cache Configuration
BACKEND_CACHE_PROVIDER	Cache provider, options: sqlite, memory, redis	sqlite	-	redis
BACKEND_CACHE_REDIS_URI	Required Redis connection URI for computed task delivery; also used for cache when BACKEND_CACHE_PROVIDER is redis	-	Yes	redis://default:teable@127.0.0.1:6379/0
Performance Cache Configuration
BACKEND_PERFORMANCE_CACHE	Performance cache Redis URL for query result caching, improves multi-user collaboration efficiency. Recommended to use separate Redis instance from BACKEND_CACHE_REDIS_URI	-	-	redis://default:teable@127.0.0.1:6379/0
Authentication Configuration
SOCIAL_AUTH_PROVIDERS	List of social auth providers, comma-separated	-	-	github,google,oidc
BACKEND_GITHUB_CLIENT_ID	GitHub OAuth client ID	-	-	github_client_id
BACKEND_GITHUB_CLIENT_SECRET	GitHub OAuth client secret	-	-	github_client_secret
BACKEND_GOOGLE_CLIENT_ID	Google OAuth client ID	-	-	google_client_id
BACKEND_GOOGLE_CLIENT_SECRET	Google OAuth client secret	-	-	google_client_secret
BACKEND_OIDC_CLIENT_ID	OIDC client ID	-	-	google_client_id
BACKEND_OIDC_CLIENT_SECRET	OIDC client secret	-	-	google_client_secret
BACKEND_OIDC_CALLBACK_URL	OIDC callback URL	-	-	https://app.teable.ai/api/auth/oidc/callback
Security & Verification
TURNSTILE_SITE_KEY	Cloudflare Turnstile site key for authentication verification	-	-	1x00000000000000000000AA
TURNSTILE_SECRET_KEY	Cloudflare Turnstile secret key for authentication verification	-	-	1x0000000000000000000000000000000AA
BACKEND_SIGNUP_VERIFICATION_CODE_RATE_LIMIT_SECONDS	Rate limit interval (seconds) for sending signup verification emails	-	-	30
BACKEND_TRUST_PROXY	Reverse proxy trust setting used to resolve the client IP behind a proxy. Affects audit logs and related security records. Accepts `true`, `false`, a hop count, or an IP/CIDR/preset list. Empty trusts private-network proxies	loopback, linklocal, uniquelocal	-	loopback, linklocal, uniquelocal
Email Configuration (Deprecated - Use Admin Panel > Instance settings > Email)
BACKEND_MAIL_HOST	Default mail server address. We recommend configuring Notify email and Automation email visually in Admin Panel > Instance settings > Email	smtp.teable.ai	-	smtp.gmail.com
BACKEND_MAIL_PORT	Default mail server port	465	-	465
BACKEND_MAIL_SECURE	Whether the default mail service uses SSL/TLS	true	-	true
BACKEND_MAIL_SENDER	Default sender address	noreply.teable.ai	-	noreply@company.com
BACKEND_MAIL_SENDER_NAME	Default sender name	Teable	-	Teable
BACKEND_MAIL_AUTH_USER	Default mail server authentication username	-	-	username
BACKEND_MAIL_AUTH_PASS	Default mail server authentication password	-	-	usertoken
BACKEND_MAIL_INVITE_USER_NAME_MAX_LENGTH	Maximum inviter name length in invitation email subjects. Empty or 0 disables truncation	-	-	32
BACKEND_MAIL_INVITE_SPACE_NAME_MAX_LENGTH	Maximum space or Base name length in invitation email subjects. Empty or 0 disables truncation	-	-	32
Session/JWT Configuration
BACKEND_SESSION_EXPIRES_IN	Session expiration time	7d	-	7d
BACKEND_SESSION_COOKIE_SECURE	Whether to secure session cookie	false	-	true
BACKEND_SESSION_ORIGIN_CHECK_ENABLED	Enable Origin and Fetch Metadata checks for unsafe browser session-cookie API requests. Enable only when your reverse proxy or CDN preserves these headers	false	-	false
BACKEND_JWT_EXPIRES_IN	JWT expiration time	20d	-	20d
BACKEND_RESET_PASSWORD_EMAIL_EXPIRES_IN	Reset password email expiration time	30m	-	30m
Resource Limits
MAX_COPY_CELLS	Maximum number of cells to copy in a single request	-	-	50000
MAX_READ_ROWS	Maximum number of rows to read in a single request	-	-	10000
MAX_ATTACHMENT_UPLOAD_SIZE	Maximum attachment upload size (bytes)	-	-	2147483648
TASK_MAX_FIELDS_PER_BATCH	Maximum number of AI fields sent to the model in one batch request. Values are clamped from 1 to 20	5	-	5
TASK_MAX_CONCURRENCY	Maximum pending AI field task runs dispatched in one scheduling cycle. Values are clamped from 1 to 20	5	-	5
TABLE_LIMIT_FIELD_OPTIONS_MAX_BYTES	Maximum serialized field option bytes	262144	-	262144
TABLE_LIMIT_SELECT_CHOICES_MAX	Maximum select choices per select field	1000	-	1000
TABLE_LIMIT_SELECT_CHOICE_NAME_MAX_LENGTH	Maximum select choice name length	1000	-	1000
TABLE_LIMIT_SELECT_DEFAULT_VALUES_MAX	Maximum select default values	100	-	100
TABLE_LIMIT_CELL_VALUE_MAX_BYTES	Maximum cell value bytes	262144	-	262144
TABLE_LIMIT_RECORD_FIELDS_MAX_BYTES	Maximum serialized record fields bytes	1048576	-	1048576
TABLE_LIMIT_RECORDS_PER_MUTATION_MAX	Maximum records per mutation	20000	-	20000
TABLE_LIMIT_COMPUTED_CELL_VALUE_MAX_BYTES	Maximum computed cell value bytes	262144	-	262144
TABLE_LIMIT_FORMULA_MAX_LENGTH	Maximum formula length	8192	-	8192
TABLE_LIMIT_TABLES_PER_BASE_MAX	Maximum tables per base	1000	-	1000
TABLE_LIMIT_FIELDS_PER_TABLE_MAX	Maximum fields per table	500	-	500
TABLE_LIMIT_VIEWS_PER_TABLE_MAX	Maximum views per table	100	-	100
TABLE_LIMIT_CREATE_TABLE_FIELDS_MAX	Maximum fields accepted when creating a table	1000	-	1000
TABLE_LIMIT_CREATE_TABLE_VIEWS_MAX	Maximum views accepted when creating a table	20	-	20
TABLE_LIMIT_CREATE_TABLE_RECORDS_MAX	Maximum records accepted when creating a table	20000	-	20000
TABLE_LIMIT_RECORDS_PER_TABLE_MAX	Maximum rows per table. Empty means no instance-level row limit	-	-	1000000
TABLE_LIMIT_VIEW_FILTER_ITEMS_MAX	Maximum filter items in a view	100	-	100
TABLE_LIMIT_VIEW_FILTER_DEPTH_MAX	Maximum nested filter depth in a view	5	-	5
TABLE_LIMIT_VIEW_SORT_ITEMS_MAX	Maximum sort items in a view	20	-	20
TABLE_LIMIT_VIEW_GROUP_ITEMS_MAX	Maximum group items in a view	3	-	3
TABLE_LIMIT_VIEW_OPTIONS_MAX_BYTES	Maximum serialized view option bytes	262144	-	262144
TABLE_LIMIT_NAME_MAX_LENGTH	Maximum display name length for supported table objects	100	-	100
TABLE_LIMIT_DESCRIPTION_MAX_LENGTH	Maximum description length for supported table objects	2000	-	2000
AI & App Builder (runtime plane connection)
SANDBOX_PROVIDER	Sandbox provider for AI sessions; `opensandbox` connects to a self-hosted runtime plane	-	-	opensandbox
TEABLE_INFRA_API_URL	Runtime plane entry URL — must be the public entry (it routes `/v1` to the sandbox engine), not an internal service address	-	-	https://infra.teable.example.com
TEABLE_INFRA_API_KEY	API key shared with the runtime plane	-	-	your-infra-api-key
TEABLE_INFRA_BUCKET	Bucket for the AI workspace data plane	teable-agent	-	teable-agent
SANDBOX_OPENSANDBOX_IMAGE	Sandbox agent image prefix without a tag — Teable appends its own release tag so the agent always matches the app version. China deployments use the Aliyun mirror prefix	-	-	ghcr.io/teableio/teable-sandbox-agent
SANDBOX_OPENSANDBOX_RUNTIME	Sandbox engine runtime type: `kubernetes` or `docker`. Must be `docker` on Docker deployments	kubernetes	-	docker
SANDBOX_OPENSANDBOX_USE_SERVER_PROXY	Route sandbox endpoints through the engine’s path-based proxy instead of per-port subdomains (used by Docker local mode)	false	-	true
APP_DEPLOY_PROVIDER	App Builder deployment backend; full-featured self-host uses `docker-runtime`	vercel	-	docker-runtime
SANDBOX_JWT_SECRET	Secret signing sandbox sessions — generate a random value per host	-	-	random string
SANDBOX_CPU	vCPUs per sandbox (admin panel settings take precedence)	2	-	2
SANDBOX_MEMORY	Memory (GiB) per sandbox (admin panel settings take precedence)	4	-	4
SANDBOX_DISK	Ephemeral disk (GiB) per sandbox (admin panel settings take precedence)	15	-	15
Feature Toggles
RECORD_HISTORY_DISABLED	Whether to disable record history, defaults to false	false	-	true
BACKEND_RECORD_HISTORY_COLD_DISABLED	Whether to stop scheduled record-history cold-storage migration, compaction, and buffer deletion. Reads still include migrated history	false	-	true
PASSWORD_LOGIN_DISABLED	Whether to disable password login (OAuth and OIDC still available)	false	-	true
Analytics & Monitoring
MICROSOFT_CLARITY_ID	Microsoft Clarity metrics ID, for enabling Microsoft Clarity analytics	-	-	your-metrics-id
OTEL_EXPORTER_OTLP_ENDPOINT	OpenTelemetry OTLP endpoint	-	-	http://jaeger:4317
TELEMETRY_REPORT_DISABLED	Disable self-hosted telemetry reporting, including license compliance reports from connected instances	false	-	true
Database Configuration
PRISMA_DATABASE_URL	Database connection URL, must be configured	-	Yes	postgresql://teable:teable@127.0.0.1:5432/teable
PUBLIC_DATABASE_PROXY	Externally reachable database address (host:port) used in the credentials generated by the base “Database Connection” panel; set it to enable external read-only SQL access	-	-	db.example.com:42345
PRISMA_TRANSACTION_TIMEOUT	Maximum time (ms) a transaction can run before timing out. Increase for long-running transactions (e.g., bulk updates with many foreign keys)	5000	-	60000
PRISMA_TRANSACTION_MAX_WAIT	Maximum time (ms) to wait to acquire a transaction from the pool	2000	-	5000
BIG_TRANSACTION_TIMEOUT	Timeout (ms) for large internal transactions, such as exporting large bases	600000	-	1200000

Self-Hosted License

Quick Start

Admin Panel

Configuration

Compliance & Telemetry

Maintenance

Deprecated

Environment Variables