APIMart

One OpenAI-compatible gateway to 190+ text, image and video models

Advanced API
Visit APIMartBack to AI tools

Own APIMart? Verify ownership to get a dofollow link

Screenshot of APIMart, One OpenAI-compatible gateway to 190+ text, image and video models

What is APIMart?

APIMart is a developer tool. A unified, OpenAI-compatible API gateway that gives developers one account, key and billing balance for 190+ chat, image and video models from OpenAI, Anthropic, Google, ByteDance and others, with health-aware routing and a shared console for logs and cost tracking.

APIMart is an API gateway that places more than 190 AI models behind a single OpenAI-compatible endpoint. Instead of holding separate accounts with OpenAI, Anthropic, Google, ByteDance, Alibaba, MiniMax, xAI and Black Forest Labs, a team connects once and reaches chat, image, video and multimodal models through the same request format. Model families named on the site include Gemini, Claude, GPT-Image, Nano Banana, Seedream, FLUX 3, Veo, Kling, MiniMax H3, Wan 3.0 and Grok Imagine. The product is aimed at developers and product teams who are shipping AI features and care about the cost and reliability of each call. Keys, credits, routing and logs live in one console, and credits are shared across model families so a balance is not stranded in a single vendor account. The gateway advertises health-aware routing across providers, which means a request can be sent to a working route when a vendor is degraded. Pricing on the model pages is quoted per image, per second of video, or per million tokens for chat, with a stated discount against official list prices and pay-only-for-successful-requests billing. Adoption is deliberately low friction. Because the interface mirrors OpenAI, existing SDK code generally needs a base URL and key change rather than a rewrite, and a playground is available for testing image and video prompts before wiring anything up. APIMart also publishes an llms.txt index that can be pasted into coding agents such as Codex, Claude or Cursor, so an assistant can answer questions about endpoints and parameters on the developer's behalf. Where it fits: this is infrastructure, not an end-user creative suite. It competes with going direct to each model vendor and with other aggregation gateways. The trade-off is familiar for any middle layer. You gain consolidation, one invoice and easier model swapping, and you accept dependence on an intermediary for uptime, pricing changes and access to the newest releases.

How do you use APIMart?

  1. 1Create an account and open the playground
    Sign in, then use the playground to send a first image or video prompt and confirm the response format before touching code.
  2. 2Generate an API key
    Open the key management page and issue a key for your project. Keep separate keys per environment so usage stays readable in the logs.
    APIMart — Generate an API key
  3. 3Pick and price your models
    Browse the model catalogue, compare per-image and per-second rates, and note the exact model IDs you plan to call.
    APIMart — Pick and price your models
  4. 4Point your existing SDK at the gateway
    Swap the base URL and key in your OpenAI SDK setup, then run a test request to a chat model to verify auth and billing.
    APIMart — Point your existing SDK at the gateway
  5. 5Load the documentation into your coding agent
    Use the agent tools page to copy a prompt that hands the llms.txt index to Codex, Claude or Cursor for help with parameters.
    APIMart — Load the documentation into your coding agent
  6. 6Review spend and call distribution
    Check the console dashboard for balance, request counts and per-model call distribution, then tune routing and budgets.

Pros and cons

Pros

  • One OpenAI-compatible endpoint and SDK path across chat, image and video modelsAI
  • Shared credit balance covers every model family, avoiding stranded prepaid balancesAI
  • Model pages list concrete per-image, per-second and per-token rates with stated discountsAI
  • llms.txt prompt lets coding agents read the docs and answer endpoint questionsAI
  • Pay only for successful requests, with routing and logs in a single consoleAI

Cons

  • Adds an intermediary layer between your product and the model vendors, so their availability becomes your availabilityAI
  • The extra 20% saving applies to selected models, not across the whole catalogueAI
  • No public terms on rate limits, quotas or service-level guarantees appear on the homepageAI
  • Routing and failover behaviour is described in marketing terms rather than documented mechanicsAI
  • Teams with one primary vendor may find a direct contract simpler than a shared credit poolAI

How much does APIMart cost?

Pricing

Usage based. Model pages quote per-image, per-second video and per-million-token chat rates, generally advertised at roughly 20% below official prices, with some image models showing higher savings (for example 44% on the Nano Banana series). Billing is pay-per-successful-request against a shared credit balance.

Learn more

Support

The homepage lists API docs, pricing, models and WeChat under one console, and frames private support, recharge and documentation as a single service flow.

Learn more

Integrations

OpenAI-compatible, so existing OpenAI SDK code can migrate with a base URL and key change. Vendor models come from OpenAI, Anthropic, Google, ByteDance, Alibaba, MiniMax, xAI and Black Forest Labs. An llms.txt index supports Codex, Claude and Cursor.

Learn more

Features

A single OpenAI-compatible gateway for 190+ models across text, image, video and multimodal work, with unified keys and credits, health-aware provider routing, per-model pricing, call and token logs, spend monitoring and a prompt-based llms.txt index for coding agents. A playground is included for testing prompts.

Learn more

Frequently asked questions about APIMart

  • How much does APIMart cost?
    Usage based. Model pages quote per-image, per-second video and per-million-token chat rates, generally advertised at roughly 20% below official prices, with some image models showing higher savings (for example 44% on the Nano Banana series). Billing is pay-per-successful-request against a shared credit balance.
  • Does APIMart have an API?
    Yes, APIMart offers an API.
  • How do you use APIMart?
    The walkthrough on this page covers 6 steps: 1. Create an account and open the playground 2. Generate an API key 3. Pick and price your models 4. Point your existing SDK at the gateway 5. Load the documentation into your coding agent 6. Review spend and call distribution.
  • What platforms does APIMart support?
    APIMart is available on Web App.
  • What does APIMart integrate with?
    OpenAI-compatible, so existing OpenAI SDK code can migrate with a base URL and key change. Vendor models come from OpenAI, Anthropic, Google, ByteDance, Alibaba, MiniMax, xAI and Black Forest Labs. An llms.txt index supports Codex, Claude and Cursor.
  • What are the limitations of APIMart?
    Adds an intermediary layer between your product and the model vendors, so their availability becomes your availability. The extra 20% saving applies to selected models, not across the whole catalogue. No public terms on rate limits, quotas or service-level guarantees appear on the homepage.

Status

StatusActive
Views0
Outbound clicks0
Added10/21/2026

Platforms

Web App

Pricing

PaidFree Trial

Categories

DeveloperGenerative ImageUnrestricted AI ChatChatCodingImage Generator