Unified API · Multi-provider access · Team controls

AI API Gateway for Teams and Businesses

Connect and manage multiple model providers through one unified API

Centralize API keys, quotas, billing, and request status. Compatible with Claude Code, Codex, Gemini CLI, and standard API workflows. Sign in to verify current models, prices, regions, and availability.

Start BuildingLearn More
PartnersEnterprise procurement & creator referrals
Business systemsERP · CRM · OAAI appsWeb · MobileAgentsAgents · WorkflowsDeveloper toolsIDE · CLI · PluginsCubence API GatewayKeyQuotaBillingPOST /v1/messagesIllustrative{"model": "<model-id>","messages": [{ "role": "user","content": "Summarise this file"}] }{ "output": "..."▍ }AnthropicOpenAIGoogleDeepSeekQwenMore modelsSee the model list for currently available models

Connecting the model and tool ecosystem

  • Anthropic
  • OpenAI
  • Gemini
  • DeepSeek
  • Qwen
  • Kimi
  • GLM
  • MiniMax
  • Claude Code
  • Codex
  • Gemini CLI
  • openCode
  • Cherry Studio
  • Cursor
  • Obsidian
  • Mastra

One model endpoint, many ways to use it

Connect team tools, agent workflows and your own business systems.

Team AI tools

Point your team's AI tools at one endpoint

Configure Claude Code, Codex, Gemini CLI and other clients with the Cubence endpoint and key, and give each member their own key and quota.

  1. 1Assign keys
  2. 2Configure clients
  3. 3Team usage
Read the API docs

Application screens are integration illustrations.

Once connected, management stays simple

Keys, quotas and spend live in one console, and every call can be traced.

Open the console
API keysSample data
New key

Total keys

3

Active

2

Disabled

1

Unlimited

1

  • Business system
    Unlimited

    sk-user-••••••••••••••••

    $132.50 used
  • Knowledge base app
    Active

    sk-user-••••••••••••••••

    $24.00 / $100.00 · 24%
  • Test environment
    Disabled

    sk-user-••••••••••••••••

    $5.00 / $50.00 · 10%

Quota limit

Set a fixed amount, or choose unlimited.

Keys and quotas per application

Issue a separate key for each system or environment, set a fixed quota or leave it unlimited, and see status at a glance.

Open the console
Cost breakdownSample data
Last 14 days

Spend this period

$30.40

  • INPUT
  • OUTPUT
  • CACHE

INPUT

$12.10

OUTPUT

$14.70

CACHE

$3.60

See the cost of every call

Input, output and cache tokens are billed item by item, with trends and details available whenever you need them.

See current pricing

GET STARTED

From setup to first call in three steps

Keep your existing code. Swap the base URL and key, and you are connected.

  1. 01

    Create an API key

    Sign in to the console, create a key for the application and set a quota if you need one.

    Business system…
    sk-user-3f9a••••••••c21eCopy
  2. 02

    Point to the API base URL

    In your client or SDK, set base_url to Cubence and paste in the key.

  3. 03

    Make a model call

    Send requests using the native Anthropic, OpenAI or Gemini protocol and get responses right away.

    POST /v1/messagesclaude-fable-5-1
    Waiting for configuration…
Create a free keyRead the docs

GROW WITH CUBENCE

Connect your business, share the value

Whether you buy for a team or share Cubence with your audience, there is a partnership track for you.

Enterprise

Enterprise procurement

Bulk purchasing, corporate billing and a dedicated contact for teams of every size.

  • Bulk procurement
  • Corporate billing
  • Dedicated support
Talk to us

How it works

  1. 1

    Submit your request

    Fill in your company details and partnership needs

  2. 2

    Discuss the plan

    Our team will reach out to discuss a plan that fits

  3. 3

    Start the partnership

    Once confirmed, sign the agreement and kick things off

Creators

Creator referrals

Get a referral link, share your AI practice and earn commissions on every referral.

  • Dedicated referral link
  • Referral commission
  • Co-created content
Apply as a creator

How it works

  1. 1

    Apply to join

    Fill in your creator profile and content focus

  2. 2

    Share your link

    Include your dedicated link in your content

  3. 3

    Earn commissions

    Get rewarded based on effective referrals

CUBENCE PRICING

Pay-as-you-go pricing

Cubence offers pay-as-you-go model access. Prices vary by model and billing item, and subscription plans are not offered.

  • Review the model identifier and its current input, output, cache, or request price
  • Charges are based on actual model usage without a monthly subscription
  • Check current service availability and regional access before use
View current model prices

Frequently Asked Questions

Everything you need to know about CUBENCE

  • What is CUBENCE?

    CUBENCE is an AI API gateway for teams and businesses. It provides unified access to multiple model providers, model selection, usage reporting, and account controls. Claude Code, Codex, and Gemini CLI are compatible client examples, not the complete model inventory.

  • How does pricing work?

    Cubence uses pay-as-you-go pricing. See the public pricing page for current model base rates, then confirm applicable group and time-based multipliers in Model Plaza. Subscription plans are not offered.

  • What models are supported?

    The public model catalog lists active model identifiers and billing modes. Confirm regional availability, client protocol and group access in Model Plaza before use.

  • Which countries and regions does CUBENCE serve?

    CUBENCE is currently available only in supported countries and regions. See the Supported Countries and Regions policy in the footer for the current list. Service availability may change based on applicable law, payment coverage, and third-party provider requirements.

  • Where can I check service availability?

    After signing in, open the Cubence status page to view the latest available component state. Cubence does not publish an uptime SLA on this page.

    Cubence | Multi-model AI API Gateway