APIMart

One AI APIConnect to top global models at a discount

Use one OpenAI-compatible API to reach leading global models across text, image, video, and multimodal workloads. APIMart unifies keys, billing, monitoring, and production access so teams can ship AI features faster at a lower operating cost.

Get StartedView DocumentationAgent ToolsNEW

Service & Support

Models, billing, API docs and support in one console.

TodayYesterdayLast 7 DaysLast 30 Days

Current Balance (Credits)

36155.97

Historical Consumption (Credits)

64.02

Request Count

165

Statistics Count

42

180%vs previous period

Statistics Quota (Credits)

18.44

256.2%vs previous period

Total Tokens

209,010

789.7%vs previous period

Spend Trend

Total: 18.44 Credits

Call Distribution

Total: 95

39
gpt-image-2-official39
claude-sonnet-4.528
veo3.1-fast16

Call Ranking

Total: 95

gpt-image-2-official39
claude-sonnet-4.528
veo3.1-fast16
seedream-4.512
OpenAIOpenAI
AnthropicAnthropic
GoogleGoogle
ByteDance
QwenQwen
KimiKimi
MinimaxMiniMax
OpenAIOpenAI
AnthropicAnthropic
GoogleGoogle
ByteDance
QwenQwen
KimiKimi
MinimaxMiniMax
OpenAIOpenAI
AnthropicAnthropic
GoogleGoogle
ByteDance
QwenQwen
KimiKimi
MinimaxMiniMax

Defining your AI API layer

Consolidate model access, cost controls, keys, monitoring, and failover into one gateway, so teams can manage every AI request through a single OpenAI-compatible interface.

01

Model Independence

OpenAIClaudeGeminiAPIMart best route

02

Sovereign Access

OpenAI Compatible

One line migration from existing SDKs

✓

Unified Billing

Credits shared by every model family

✓

Provider Routing

Health-aware routing across vendors

✓

Private Support

Docs, recharge, and service in one flow

✓

OpenAI · Anthropic · Google · ByteDance — one account, one invoice

03

Across the API lifecycle

SHARED CONTROL
KEYS
PRICE
ROUTE
CALL
TASK
BILL
LOGS
SUPPORT

Not just access. Every request strengthens cost, routing, and visibility.

llms.txt

One prompt — use APIMart from any Agent

Copy a single prompt to CodeX, Claude, Cursor or any Agent so it instantly understands every model and endpoint on APIMart, and ask it your questions.

apimart — llms.txt prompt
Please read APIMart's documentation index below and answer my questions based on it:

https://docs.apimart.ai/llms.txt

About APIMart: a unified gateway for 190+ AI models — chat (GPT, Claude, Gemini), image (Nano Banana, GPT-Image-2, Imagen), video (Veo, Kling, Seedance) and audio — all through an OpenAI-compatible API. Pay only for successful requests.

My question: How do I generate a 5-second video with Kling V3 in Python?

Models

Popular Models

Curated top models across categories — start building in minutes.

View all models

3 minute quick integration

Within minutes, you can start using hundreds of AI models.

  1. 1
    Sign Up & Create KeyRegister an account and generate your API key in the dashboard — signing up is free.
    Get API KeyTry Online
  2. 2
    Update ConfigurationChange the API endpoint to APIMart — OpenAI SDK users only need to modify one line of code.
    View DocsTry Online
  3. 3
    Start Calling ModelsAccess 190+ AI models like GPT-6.1, Claude and Veo through one unified API.
apimart — demo.request
curl --request POST \
  --url https://api.apimart.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gpt-4o",
    "messages": [
      {
        "role": "system",
        "content": "You are a professional AI assistant."
      },
      {
        "role": "user",
        "content": "Tell me about the history of artificial intelligence."
      }
    ]
  }'

Why APIMart

Why teams choose APIMart

One platform, clear pricing, real support — so you can focus on shipping.

FAQ

FAQ

What is APIMart and what does it offer?

APIMart is a unified AI API platform providing access to 190+ AI models through a single API. Access Chat APIs (GPT-6.1, Claude Opus 5.5, Gemini 3.8), Video APIs (Veo 3.1, Seedance 2.5) and Image APIs (GPT Image 2, Nano Banana Pro) with OpenAI-compatible integration, priced 20% below official rates.

How does pricing compare to other API providers?

Every model on APIMart is priced 20% below the official provider rate, and Platinum and Diamond membership tiers save up to 28%. Pay-as-you-go pricing with no monthly minimums or hidden fees, and you only pay for successful requests.

How do I integrate the APIMart API into my application?

Simple integration: 1) Sign up and create an API key, 2) Update your base_url to APIMart's endpoint, 3) Start using the OpenAI-compatible format. For OpenAI users, it's just a one-line code change with comprehensive documentation and examples.

Why should I choose APIMart over using model providers directly?

With APIMart, you only need to manage one account and one invoice to access 190+ AI models. No more maintaining separate API keys for Claude, OpenAI and Google. Plus, every model is priced 20% below the official rate.

Is my API key secure?

Your keys are encrypted in our database. You can generate new keys or revoke old ones at any time in the console to keep your account secure.

Which payment methods are supported? Can I get an invoice?

APIMart supports multiple mainstream payment channels: Antom (Visa, Mastercard, Alipay), Crypto, PayPal, and Creem. After selecting a plan in the console, you can pay through the channels available in your region. All top-up records can be queried and exported from the billing page. For official electronic invoices or corporate bank transfers, please contact our support team for assistance.

Which programming languages are supported?

APIMart is fully compatible with the OpenAI API specification, so the official OpenAI client libraries (Python, Node.js, Go, Java, PHP, C#, Ruby, etc.) work out of the box—just swap the base_url, no extra dependencies required. We also support mainstream frameworks like LangChain, LlamaIndex, and Vercel AI.

Are there rate limits? How is stability ensured under high concurrency?

Default accounts come with generous RPM/TPM quotas, and enterprise accounts can request dedicated quotas and isolated channels. APIMart aggregates multiple upstream providers behind one unified API, using multi-region, multi-provider automatic routing and failover to reduce single-point risks. Failed requests are not billed, and you can review every request and your usage in the console's request logs.

Will my requests and data be used to train models?

No. APIMart only forwards your request to the selected model provider for the current inference; we never retain your inputs or outputs to train any model. We also pass through each provider's zero data retention (ZDR) agreements, and enterprise accounts can sign a DPA to meet GDPR, SOC 2, and other compliance requirements.

Get started

Ready to get started?

Trusted by developers · Powered by leading AI providers

Get startedView documentation