Modal logo
Introduction Functions Functions Invocation methods Function lookups Container lifecycle management Parametrized Functions Dynamic configuration Custom container images Defining Images Using existing container images Named images Fast pull from registry GPUs and other resources GPU acceleration Using CUDA on Modal Configuring CPU, memory, and disk Scaling out Scaling out Input concurrency Batch processing Job queues Dynamic batching Multi-node clusters BetaDeployment Apps, Functions, and entrypoints Managing deployments Continuous deployment Scheduled Functions Endpoints Endpoints Endpoint metrics Benchmark an endpoint Endpoint integrations Sandboxes Sandboxes Running commands Networking and security File access Snapshots Pricing and resources VM Sandboxes BetaNotebooks Secrets and environment variables Secrets Environment variables HTTP Applications Servers Web Functions Streaming endpoints Web Function URLs Request timeouts Proxy Tokens Networking Tunnels Proxies BetaCluster networking Data sharing and storage Passing local data Volumes Storing model weights Cloud bucket mounts Dicts Queues Dataset ingestion Performance Cold start performance Memory Snapshots High-performance LLM inference Reliability and robustness Failures and retries Preemption Timeouts GPU health Troubleshooting Security and privacy Security and privacy Customer-supplied encryption keys AlphaAudit logs Integrations Using OIDC to authenticate with external services Connecting Modal to your Datadog account Connecting Modal to your OpenTelemetry provider Okta SSO Custom SAML SSO SCIM Integration BetaSlack notifications BetaWorkspace & account settings Workspaces Environments Modal user account setup Service users Role-Based Access Control (RBAC) Billing Budgets Other topics Feature maturity Restricted Functions JavaScript/Go SDKs BetaModal 1.0 migration guide File and project structure Developing and debugging Jupyter notebooks Asynchronous API usage Global variables Region selection S3 Gateway endpoints GPU Metrics

Proxy Tokens

Use Proxy Tokens to prevent unauthorized clients from reaching your Endpoints, Servers, and Web Functions. Proxy Tokens can be created in the Dashboard settings or with the modal workspace proxy-tokens CLI.

Authentication models 

Endpoints and Servers require authentication by default. To accept public traffic instead, pass --unauthenticated to modal endpoint create or set unauthenticated=True in the @app.server() decorator:

@app.server()
class Private:
    ...


@app.server(unauthenticated=True)
class Public:
    ...

In contrast, Web Functions are publicly available by default. Enable authentication by setting requires_proxy_auth=True in the fastapi_endpoint, asgi_app, wsgi_app, or web_server decorators:

@app.function()
@modal.fastapi_endpoint()
def public():
    return "hello world"


@app.function()
@modal.fastapi_endpoint(requires_proxy_auth=True)
def private():
    return "hello friend"

The public endpoint can be hit by any client over the Internet:

curl https://public-url--goes-here.modal.run

The private endpoint cannot:

curl --fail-with-body https://private-url--goes-here.modal.run
# modal-http: missing credentials for proxy authorization
# curl: (22) The requested URL returned error: 401
# https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/401

Authenticating requests 

A Proxy Token comprises a Token ID / Token Secret pair. Requests are authenticated by passing the pair in HTTP headers. They can be sent as separate Modal-Key and Modal-Secret headers:

export TOKEN_ID=wk-1234abcd
export TOKEN_SECRET=ws-1234abcd
curl -H "Modal-Key: $TOKEN_ID" \
     -H "Modal-Secret: $TOKEN_SECRET" \
     https://private-url--goes-here.modal.run

Alternatively, they can be joined with a period (.) and passed as a single Authorization: Bearer header:

export TOKEN_ID=wk-1234abcd
export TOKEN_SECRET=ws-1234abcd
curl -H "Authorization: Bearer $TOKEN_ID.$TOKEN_SECRET" \
     https://private-url--goes-here.modal.run

This is the same scheme the OpenAI API uses (Authorization: Bearer <api-key>), so the combined value can be used as the API key in any OpenAI-compatible client or gateway.

Environment scoping 

On Workspaces with RBAC enabled, tokens are scoped to specific Environments. See the RBAC guide for more information.