Question 1

What is BabySea?

Accepted Answer

BabySea is the execution control plane for generative media. It gives developers one unified API for AI image generation and AI video generation across inference providers. BabySea manages provider routing, failover, execution state, billing, signed webhooks, observability, and lifecycle controls so teams can build against one predictable execution system instead of maintaining separate provider integrations. Learn more on the why we built BabySea.

Question 2

How do I get started?

Accepted Answer

Sign up for free with no card required. Every new account starts with $1 in welcome credits. Create an API key, verify it with `GET /v1/status`, then send your first image or video generation request. The quickstart guide walks through the basic flow, and the playground lets you test live requests from your browser.

Question 3

How long does generation take?

Accepted Answer

Image jobs usually finish in seconds, while video jobs take longer depending on the model, provider, duration, resolution, and output settings. When multiple providers are available for a model, BabySea can use your configured provider order or `fastest` routing to prefer faster and healthier provider paths.

Question 4

What is the unified schema?

Accepted Answer

BabySea exposes one execution schema across providers. Instead of adapting your application to provider-specific inputs, enum names, safety flags, file conventions, or output formats, you send a normalized BabySea request. BabySea validates the request, maps it to the selected provider, and keeps output behavior consistent. See the model schema for model-specific fields.

Question 5

What providers does BabySea support?

Accepted Answer

BabySea runs workloads across multiple inference providers. Provider availability can vary by model and region, so use the current model pricing and model schema before building against a model. Your application can use one execution API instead of separate provider SDKs, schemas, webhook formats, and failover logic.

Question 6

What image models are available?

Accepted Answer

The image catalog covers fast drafts, high-resolution generation, photorealism, text rendering, editing, and provider-specific model families. Model availability, supported fields, and pricing can vary by provider and region. Use the model pricing and model schema to compare current image models before building against them.

Question 7

What video models are available?

Accepted Answer

The video catalog includes models for short clips, cinematic output, image-to-video workflows, and provider-specific video generation capabilities. Video usage is usually priced by output duration and can also vary by model, resolution, and provider. Check the model pricing for current video model costs.

Question 8

How do I choose the right model?

Accepted Answer

Choose based on output type, supported fields, latency, provider coverage, quality profile, and cost. Compare models on the model pricing, inspect schemas on the model schema, then call `GET /v1/estimate/{model_identifier}` to preview cost before running a generation.

Question 9

What do I get for free?

Accepted Answer

Every new user receives $1 in welcome credits automatically. The free plan is designed for testing the API, creating keys, sending generation requests, receiving signed webhooks, and validating the execution lifecycle before upgrading. See the pricing for the current plan limits and included features.

Question 10

How does pricing work?

Accepted Answer

BabySea uses a subscription plus credits model. Each plan includes monthly credits, feature limits, and retention windows. Successful generations deduct from your credit balance based on model, output type, duration, resolution, and request parameters. You can preview estimated cost with `GET /v1/estimate/{model_identifier}`. See billing plans and credit packs guide.

Question 11

What credit packs are available?

Accepted Answer

Accounts with an active subscription can buy one-time credit packs when they need more usage beyond their monthly allocation. Purchased credits are added to the account balance and can be used for future generations. Only authorized account roles can complete billing actions. See the credit packs guide.

Question 12

Do credits expire?

Accepted Answer

Credits remain on your balance until they are used or adjusted through a billing operation. Every reserve, charge, refund, and credit addition is written to the credit ledger, so the balance remains auditable over time. You can review credit movement from the dashboard or through the credit transactions.

Question 13

How much does each generation cost?

Accepted Answer

Cost depends on the model, output type, and settings. Image models are usually priced per generation, while video models are often priced by output duration and can vary by resolution or model family. Use `GET /v1/estimate/{model_identifier}` to preview the cost before executing. Review model pricing.

Question 14

How do I integrate with the API?

Accepted Answer

You can integrate through the TypeScript SDK with `npm install babysea` or call the REST API directly. Either path gives you one execution API for generation, cost estimates, content retrieval, cancellation, webhooks, usage, billing, and health checks. Start with the SDK overview or docs.

Question 15

What can I do in the Playground?

Accepted Answer

The Playground is a browser-based API console for live generation and generation management. You can choose image or video models, inspect schemas, keep your API key in memory, send requests, view responses, copy code, cancel or delete generations, and test the same execution control plane used by the API. See the playground guide.

Question 16

How do I verify my API key and monitor usage programmatically?

Accepted Answer

Use `GET /v1/status` to smoke-test an API key and inspect metadata like key prefix, creation time, and expiry. Use `GET /v1/usage` to retrieve customer-facing usage summaries, including request counts, errors, endpoint activity, and provider cost summaries. See API keys and credit usage.

Question 17

How do API keys work?

Accepted Answer

BabySea API keys use the `bye_` prefix and authenticate public `/v1` API requests. Full key values are shown once at creation, then stored hashed at rest. Keys can have scopes, expiration, IP allowlists, and rotation windows. See API keys.

Question 18

What API key scopes are available?

Accepted Answer

API keys can be created with presets such as Full Access, Generate Only, Read Only, or Monitor Only. These map to route-level permissions across generation, account, billing, health, usage, model catalog, and content APIs. Scope checks are enforced server-side on authenticated routes. See the API scopes.

Question 19

How do webhooks work?

Accepted Answer

BabySea emits signed generation events such as `generation.started`, `generation.completed`, `generation.failed`, and `generation.canceled`. Webhooks include delivery IDs, signatures, retries, and delivery logs. See the webhook overview.

Question 20

Should I poll for results or use webhooks?

Accepted Answer

Both are supported. Polling is simplest while prototyping: call `GET /v1/content/{generation_id}` until the generation reaches a terminal state. For production systems, webhooks are usually better because BabySea can push signed lifecycle events to your endpoint as the workload progresses.

Question 21

How do I verify webhook signatures and operate webhook endpoints?

Accepted Answer

Every webhook request includes BabySea headers such as a delivery ID, event type, timestamp, and HMAC-SHA256 signature. Verify the signature with your webhook signing secret and reject stale timestamps to prevent replay. Delivery attempts are logged. Generation lifecycle deliveries retry automatically, while test, replay, and credit alert deliveries are one-shot. See signature verification and retry and delivery.

Question 22

What regions are available?

Accepted Answer

BabySea provides regional API endpoints for the US, EU, and Japan deployments. Each region has a separate API hostname and operational boundary. Choose the region that best matches your users, latency requirements, and data residency needs. See the API and inference terms.

Question 23

How do custom API domains work?

Accepted Answer

Custom API domains let Enterprise accounts expose BabySea under their own domain while keeping the default BabySea endpoint available. The lifecycle includes domain setup, verification, SSL provisioning, and safe cleanup if the account is deleted. See domains, custom domain setup, and domain lifecycle.

Question 24

How does failover routing work?

Accepted Answer

BabySea executes every workload through a provider path. If one provider times out or returns a mapped provider failure, BabySea moves the workload to the next available provider in the order. This gives your application provider failover without requiring separate retry trees, provider-specific cancellation, or error normalization.

Question 25

What does generation_provider_order: "fastest" do?

Accepted Answer

`generation_provider_order: "fastest"` lets BabySea choose provider order from current regional performance and health data. If no performance signal is available for the model, BabySea falls back to a safe supported provider order.

Question 26

How do provider timeouts work?

Accepted Answer

Each provider attempt runs inside an enforced timeout window. Image routes use shorter timeouts, while video routes allow longer execution windows because video models can take significantly more time. When a provider exceeds its window, BabySea can cancel the provider job where supported and move to the next provider.

Question 27

What happens if a provider fails?

Accepted Answer

BabySea automatically moves to the next provider in the execution path when the current provider times out or returns a mapped provider error. If every configured provider fails, the API returns a structured BabySea error with a request ID, retryability signal, and normalized provider failure details. See error reference.

Question 28

What happens if generation fails?

Accepted Answer

Reserved credits are refunded automatically for failed generations. BabySea records the failure state, normalizes the error response, and emits webhook events when configured. Pending or processing generations can also be canceled, although very fast jobs may finish before cancellation is applied.

Question 29

Can I cancel or delete generations after I submit them?

Accepted Answer

Yes. Eligible pending or processing generations can be canceled while the cancel window remains open with `POST /v1/content/generation/cancel/{generation_id}`, and the generation record and stored output files can be deleted with `DELETE /v1/content/{generation_id}`. Deletion is irreversible. See the generation lifecycle guide for lifecycle actions.

Question 30

How is my data secured?

Accepted Answer

BabySea includes scoped API keys, bcrypt-hashed secrets, per-key IP allowlists, HMAC-signed webhooks, account-level access controls, audit logging, security headers, and operational monitoring. Review the trust center and privacy policy for more detail.

Question 31

What are the rate limits?

Accepted Answer

Rate limits are enforced per account in shared sliding windows across all API keys on that account. BabySea separates general API traffic from generation traffic and scales limits by plan. Responses include standard rate-limit headers so clients can observe remaining capacity and back off cleanly.

Question 32

What analytics are available?

Accepted Answer

The dashboard covers plan and credit usage, generation outcomes, model and provider usage, request and rejection counts, webhook delivery performance, latency percentiles, provider behavior, and account activity. BabySea also exposes usage data through the API for programmatic monitoring.

Question 33

Can I export operational logs and credit history?

Accepted Answer

Yes. CSV and JSON exports are available for operational areas such as activity and logs, while the credit ledger keeps a transaction trail across reserves, charges, refunds, and credit additions. See log export, log retention, and credit transactions.

Question 34

Do you support team accounts?

Accepted Answer

Yes. Team workspaces support role-based access with `primary-owner`, `owner`, and `member` roles. Primary owners control billing and ownership responsibilities, owners can manage most operational features, and members have limited access for safer collaboration. See member roles.

Question 35

Can I purchase more credits anytime?

Accepted Answer

Yes. Accounts with an active subscription can purchase one-time credit packs from Billing when they need more execution capacity. Credit purchases are recorded in the credit ledger and become available for future workloads. See credit packs guide.

Question 36

How long are generated files stored?

Accepted Answer

Generated output files are temporary by design. Retention depends on your plan, and files are automatically deleted after the retention window. BabySea keeps the generation record and audit state separate from temporary storage so the workload history remains understandable after files expire. See data lifecycle terms.

Question 37

How long are generation logs kept?

Accepted Answer

Generation logs record request state such as model, provider, status, timing, and cost. Retention depends on plan and operational policy. Logs help customers inspect workload behavior while allowing BabySea to minimize long-term storage of generated assets and unnecessary operational data.

Question 38

How long are API request logs kept?

Accepted Answer

API request logs capture endpoint, method, status, request metadata, and error behavior for observability and troubleshooting. Retention is plan-based and documented in the logs retention guide. See log retention.

Question 39

How long are webhook event logs kept?

Accepted Answer

Webhook delivery logs track event type, destination, delivery attempts, response status, and errors. They are retained according to plan and operational policy so teams can debug delivery behavior, replay eligible deliveries, and audit endpoint reliability.

Question 40

How long are activity logs kept?

Accepted Answer

Activity logs capture account-level changes such as API key events, webhook updates, billing changes, allowlist edits, and team actions. Retention follows plan and compliance policy. See activity retention and export for details.

Question 41

How long is the credit audit trail kept?

Accepted Answer

The credit audit trail records deposits, purchases, reserves, charges, refunds, and adjustments. It is retained longer than ordinary operational logs because it supports billing reconciliation, customer support, and account-level auditability. See credit transactions.

Question 42

Can I export or delete my data?

Accepted Answer

Yes. Account data can be exported from supported dashboard areas, and generated content can be removed through the content API. Account deletion flows clean up account-owned resources while preserving minimal audit records where required. See account access and deletion and data lifecycle terms.

Question 43

How does BabySea handle data privacy?

Accepted Answer

BabySea records consent decisions, exposes privacy and cookie policy pages, supports account deletion workflows, scrubs sensitive activity metadata where appropriate, and gates analytics behavior according to consent. Review the privacy policy and trust center.

Question 44

Where can I find legal, policy, and service information?

Accepted Answer

Product documentation lives on the docs, while legal and policy documents live on the terms site. That includes terms of use, privacy policy, service terms, cookie policy, and related account or billing policies.

Question 45

Where can I review BabySea's security and compliance posture?

Accepted Answer

The trust center provides the high-level view of BabySea's security controls, infrastructure, privacy posture, and legal documentation. For implementation details, the documentation covers API security, webhooks, data lifecycle, billing, and account controls.

Question 46

What payment methods are accepted?

Accepted Answer

Billing runs through a hosted checkout and billing portal. Available payment methods depend on the billing provider, region, and account configuration shown during checkout. Subscriptions can be purchased monthly or yearly when plan options are available. See billing overview.

Question 47

Can I cancel or change my plan?

Accepted Answer

Yes. Authorized billing roles can use the billing portal to update payment details, switch plans, download invoices, or cancel a subscription. Role permissions determine who can view billing and who can perform purchases or plan changes. See billing plans.

Question 48

Can I get a refund?

Accepted Answer

Failed generations are refunded automatically through the credit ledger. For subscription disputes, billing corrections, or refund requests outside automatic generation refunds, review the billing and credit terms or contact contact us.

Question 49

Does BabySea filter or moderate content?

Accepted Answer

Content moderation is primarily enforced by inference providers. BabySea preserves provider safety behavior and returns structured errors when content is rejected instead of silently changing the request. Customers are responsible for using the platform according to applicable terms, laws, and provider safety requirements. See AI principles.

Question 50

What output formats are supported?

Accepted Answer

Output formats depend on the selected model and provider. Image models commonly return PNG, JPEG, or other image formats depending on schema support, while video models commonly return MP4. BabySea normalizes output delivery so generated content can be retrieved through the same content lifecycle APIs.

Question 51

What's on the roadmap?

Accepted Answer

We do not publish a fixed public roadmap in the FAQ. Follow the docs changelog for shipped updates, and contact support if you need a specific model, region, provider path, or account capability.