STSuperTomorrow

API-FIRST MODEL ACCESS

One API for global AI models.

Unified endpoint, authentication, metering, billing, and request logs for manageable, traceable, production-ready model access.

Each key can access multiple models. Successful requests are metered; failed requests do not consume Credits.

Usage billing

Separate input/output token rates

Key model scope

Explicit single or multi-model access

Unified API

OpenAI-compatible

POST /v1/chat/completions
curl https://api.supertomorrow.cc/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

ACCESS FLOW

A clear path from integration to production calls.

Model prices are shown per 1M tokens, while actual usage deducts Credits by input and output tokens. Create a key, bind models, add balance, and start sending requests.

1

Sign in

Manage your account with email and password.

2

Add Credits

Top-ups issue Credits for actual input/output token usage.

3

Create API Key

Bind one or multiple models to each key.

4

Select models

Choose models by capability and 1M-token pricing.

5

Send first request

Send requests through an OpenAI-compatible endpoint.

PRICE PREVIEW

Usage-based pricing with transparent rates.

Input and output are priced separately, shown per 1M tokens in USD and Credits.

Loading platform prices...

View all models
Model
Input price
Output price
Recommended use
Try
deepseek-chat
-- / 1M tokens

--

-- / 1M tokens

--

General Chinese tasks and development tests
deepseek-reasoner
-- / 1M tokens

--

-- / 1M tokens

--

Reasoning and multi-step analysis
Note: prices are SuperTomorrow platform rates. Credits are estimated at 1 USD = 100 Credits. API calls deduct by actual input/output tokens; failed requests are not charged.

TRUST & BILLING

Only successful calls are billed.

Credits act as the unified balance unit. Every successful request generates token usage, Credit deduction, model route, and request log records.

No charge for failed requests

Provider timeouts, 502/429 responses, channel errors, and failed requests are logged without creating charge transactions.

Traceable request chain

Successful calls record user, API key prefix, model, provider, tokens, Credit deduction, and request time.

Clean route policy

Commercial models use explicit provider channels and account pools; experimental channels are excluded from the official catalog.

Explicit key scoping

Each API key can be bound to one or multiple models. Unauthorized models are rejected by the backend.

SECURITY COMMITMENT

Isolation and traceability are built into the platform.

From ingress and keys to request logs and billing records, the platform is designed for secure commercial model distribution.

ST

Cloudflare edge entry

The site and API entry points are served through Cloudflare with edge ingress and TLS transport protection.

ST

No prompt body storage

The platform does not persist prompts or model response bodies; only billing, debugging, and audit metadata is retained.

ST

Isolated API keys

Each user and API key is managed independently, with model scopes, disabling, regeneration, and request tracing.

ST

Traceable billing chain

Every successful call records model, provider, tokens, Credit deduction, latency, and request status.

CONTROL SURFACE

Built for multi-model access, metering, and distribution.

Open portal

Model catalog

Models are presented by capability, price, and availability for scenario-based selection.

Key scope

Each key can be bound to one or multiple enabled models to avoid accidental access.

Wallet billing

Successful requests are charged by actual tokens. Failed requests are not charged.

Channel stability

Model channels use health checks and failover to reduce single-channel impact.

Usage records

Dashboard shows requests, tokens, spend, and balance transactions.

Status visibility

Model availability, latency, and recent request status are visible.

QUICK START

Commercial access flow

STEP 1

Register or sign in

Use email and password to enter Dashboard.

STEP 2

Add Credits

Top-ups issue Credits for usage-based billing.

STEP 3

Create API Key

Create credentials; the complete key is shown once.

STEP 4

Choose models

Select one or more available models for the key.

STEP 5

First API call

Call a model explicitly bound to your key.

API ENDPOINT

Copy endpoint

POST https://api.supertomorrow.cc/v1/chat/completions

CODE EXAMPLE

Copy ready-to-run code

Replace the API key and run the request.

curl https://api.supertomorrow.cc/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

5 MINUTES

Make a real API call in 5 minutes.

The flow from signup, top-up, key creation, and first request is built around an OpenAI-compatible API. The API endpoint remains stable even when the console domain changes later.

1

Sign in

Manage your account with email and password.

2

Add Credits

Top-ups issue Credits for actual input/output token usage.

3

Create API Key

Bind one or multiple models to each key.

4

Select models

Choose models by capability and 1M-token pricing.

5

Send first request

Send requests through an OpenAI-compatible endpoint.

curl https://api.supertomorrow.cc/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello"}],
    "max_tokens": 160
  }'

USE CASES

For teams that need unified model access and cost control.

Development testing

Validate model quality, latency, and cost across providers through one endpoint.

Multi-model cost control

Break down price, usage, and spend by model to control project-level budgets.

Team key management

Create separate keys for projects or environments, with explicit model access scopes.

Commercial distribution

Support top-ups, usage-based billing, balance transactions, and traceable call history.

BOUNDARIES

Clear boundaries for access, billing, and call responsibility.

  • This is not a chatbot and not a model marketplace.
  • Dashboard billing records and balance transactions provide the authoritative charge history.
  • Full API keys are shown only once when created or regenerated.
  • Model-channel failures, timeouts, 502/429 responses, and failed requests are not charged.

FAQ

Frequently asked questions

These notes cover model sourcing, Credit billing, failed requests, key safety, and enterprise access before integration.

Is SuperTomorrow an official model provider?+

No. SuperTomorrow is a model access, metering, and billing platform. It provides unified API access, key scopes, Credit billing, and request logs; underlying models are served by their providers.

How are Credits charged?+

Top-ups issue balance at the reference rate of 1 USD = 100 Credits. Successful calls deduct Credits by actual input tokens, output tokens, and model pricing.

Are failed requests charged?+

No. Provider timeouts, upstream errors, rate limits, or failed requests are logged without creating charge transactions.

What should I do if an API key leaks?+

Disable or regenerate the key in Dashboard. Use separate keys for different projects and bind only the required model scope.

Which models are supported?+

The Models page lists currently available production models. The platform prioritizes mainstream, stable, and traceable model routes.

Do you support refunds?+

Top-ups and usage are governed by wallet and billing records. Refunds for unused balance follow the refund policy; consumed Credits cannot be reversed.

Do you store request content?+

The platform does not persist prompts or model response bodies. It keeps request metadata required for billing, troubleshooting, and audit trails.

Do you support enterprise customers?+

Enterprise customers can manage usage by project, key, model scope, and billing records. Contact the platform for higher limits or reconciliation support.

NEXT STEP

Create an API key and send your first model request.

After signup, users can create API keys, inspect model prices, submit top-up requests, and test calls in the dashboard.