Hugging Face Inference Endpoints vs Modal

Hugging Face Inference Endpoints

OAuth scopes separate reading endpoints from writing them, both OpenAPI documents are public, and the unauthenticated /v2/provider route lists every instance with its hourly price. An account needs a payment method and credits before the first deployment, no rate limits or SLA were found for the management API, and the docs price table disagrees with the live list in places.

Modal

Scale to zero by default, per-second billing and about one-second container boots. No REST API or OpenAPI spec for deploying or invoking Functions.

Hugging Face Inference Endpoints

1. Call GET https://api.endpoints.huggingface.cloud/v2/provider first and pick an instance whose status is available. The docs table lists types the API marks deprecated or not available
2. Send X-Scale-Up-Timeout: 600 on requests to an endpoint that scales to zero, or handle 503 while the first replica starts
3. Set scaleToZeroTimeout yourself. The docs give a default of 1 hour and the OpenAPI document says 15 minutes
4. Pause or delete an endpoint when the job is done. Billing covers every minute a replica is initialising or running
5. Give the agent a fine-grained token or the read-endpoints scope unless it must deploy. Endpoints are private by default and take the same Hugging Face token as a bearer

Modal

1. Create a proxy token and require it on every web endpoint before sharing the URL; endpoints are public by default
2. Pass a list to gpu= (for example ["H100", "A100-80GB"]) so a job still runs when the first choice is unavailable
3. Set scaledown_window and min_containers explicitly; the defaults are 60 seconds and 0
4. Use .spawn() and poll the call ID for long work instead of holding a web request open
5. Keep web endpoint traffic under 200 requests a second or ask Modal to raise the limit