Log inSubmit server
Apimodels logo

Apimodels

apimodels connects MCP clients to chat, image, video, and MiniMax speech models through one API key.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Apimodels

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "apimodels": {
      "command": "npx",
      "args": [
        "-y",
        "apimodels-mcp"
      ],
      "env": {
        "APIMODELS_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

Automated check passed— started and listed 8 tools correctly (13d ago).

Overview

The apimodels MCP server connects MCP clients to apimodels.app for chat, image generation and editing, video generation, image review, and MiniMax text-to-speech. It exposes model discovery, balance checks, asynchronous task polling, and hosted file uploads alongside generation tools. Use the local server when agents need local file access, or the hosted Streamable HTTP endpoint when you prefer OAuth and no installation. It fits workflows that need an agent to create media, inspect generated images, revise prompts, and track longer-running renders.

Use cases

  • Generate images and inspect their previews
  • Create videos from prompts or reference images
  • Review generated images and revise prompts
  • Convert scripts into MiniMax speech audio
  • Poll long-running media generation tasks

Key features

  • Chat and text completion across available LLMs
  • Text-to-image and image editing
  • Text-to-video with optional reference images
  • Vision-based image review and prompt revision
  • MiniMax text-to-speech with audio URLs
  • Asynchronous task polling and hosted uploads

Tools8

What Apimodels lets your agent do. Expand a tool for its parameters.

Verified live~2.6k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

9/9 checks healthy over the last 44d

What it can do

From the 8 tools this server returned when we ran it; 8 declare their own safety hints.

Handles credentials · 1Makes network requests · 6Read-only tools · 8
Network at startup3 connections
Files written at startupNone
Install scriptsNone
npm provenanceNot published
Package age41 days
Maintainers1

Observed 2026-10-30 in an isolated sandbox. A description of reach, not a security audit.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 13d ago via OSV.dev · apimodels-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars1Stargazers on the source repository.
npm downloads1.5kPackage downloads in the last 30 days.
Last commit3d agoMost recent push to the default branch.
Availability100%Our rolling endpoint + install checks that succeeded.
Install checkPassedOur sandbox started it and listed its tools.
Tools exposed8Callable tools this server registers over MCP.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Apimodels works

What apimodels MCP server does

The apimodels MCP server gives an MCP-compatible agent access to models available through apimodels.app. A single API key can be used for LLM chat, text-to-image generation and image editing, text-to-video generation, image critique, and MiniMax text-to-speech. The server also exposes the account balance and the currently available model identifiers, grouped across chat, image, video, and audio use cases.

Image generation returns image URLs and can include a reduced preview for the model to inspect. The image workflow supports text prompts and source images. With the documented Seedream model, callers can request transparent backgrounds or decompose an image into a base image and as many as 16 named transparent layers. Video generation accepts an optional reference image. Speech generation returns an audio URL; ElevenLabs streaming is not part of this MCP interface.

How it works

The apimodels MCP server can run locally through the apimodels-mcp package or remotely at https://api.apimodels.app/mcp using Streamable HTTP. Local execution allows the server to find local paths, localhost URLs, and data image URIs, upload them, and pass the resulting public URL to the generation service. Hosted execution cannot read the client machine directly, so it provides get_upload_url, a temporary upload link that can be used with curl -T.

Generations are asynchronous. generate_image, generate_video, and text_to_speech may return a task identifier when work is still running; get_task checks the task and returns the resulting URLs when ready. The default wait period is 50 seconds, and APIMODELS_WAIT_SECONDS can change the server default. This approach avoids resubmitting a task when a client timeout occurs. Hosted calls wait no longer than 50 seconds before returning a task identifier.

Image results can be reviewed with review_image. The vision model reports issues against a supplied brief and suggests a revised prompt, allowing an agent to iterate on generated output. Clients that pass image tool results to the model can also inspect the returned preview directly.

Setup and configuration

For a local installation, run npx -y apimodels-mcp login and authorize in the browser, or create an API key in the apimodels console and provide it as APIMODELS_API_KEY. The login command stores credentials in ~/.apimodels/credentials.json, which lets configured clients use the same local credential. npx -y apimodels-mcp status reports the active key and balance. Node.js 18 or later is required for this setup.

The hosted endpoint requires no local installation. Add its URL to a client that supports remote MCP connections, then complete the apimodels.app authorization flow. A plain API key can also be sent as a Bearer token. Hosted authentication creates a dedicated API key that can be deleted from the apimodels console to disconnect the client.

Tools and capabilities

  • get_balance checks the account balance in USD.
  • list_models returns available chat, image, video, and audio model IDs.
  • chat sends chat or text-completion requests to an LLM.
  • generate_image creates or edits images and can return previews.
  • review_image critiques an image and proposes a revised prompt.
  • generate_video creates videos from text, optionally using a reference image.
  • get_task waits for or checks an unfinished generation.
  • get_upload_url supplies a temporary hosted upload URL.
  • text_to_speech creates MiniMax speech audio and returns its URL.

Limitations and notes

The hosted server cannot access local files without an explicit upload step. Local uploads are stored in the account's R2 space and automatically deleted after 30 days. Media generation consumes account credits; the Codex plugin requests approval before credit-spending calls, while balance checks, model listing, and progress checks do not require that approval.

Long video renders can take several minutes, so agents must follow task IDs with get_task rather than assuming one tool call will return the finished video. The README reports a median video duration of about 2.5 minutes and images taking about 50 seconds, based on the stated production measurements through 2026-09-22. The local server handles local paths, but the hosted endpoint requires public URLs or its upload flow.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Apimodels

apimodels is an MCP server that connects MCP clients to apimodels.app and exposes chat, image generation, image review, video generation, text-to-speech, model listing, balance checks, file uploads, and task polling. It supports local execution with an API key or a hosted Streamable HTTP endpoint with OAuth.

Install it locally by running `npx -y apimodels-mcp login`, authorizing in the browser, and adding the `npx -y apimodels-mcp` command to your MCP client. Node.js 18 or later is required. You can instead add the hosted Streamable HTTP URL without installing Node.js.

Yes, apimodels works with Cursor through the local MCP server or the hosted remote MCP URL. For local use, authenticate with the login command and configure the server. For hosted use, add `https://api.apimodels.app/mcp` as a custom connector and complete authorization.

It returns a task ID when generation is still running, then the agent calls `get_task` to check progress and retrieve the video URL. Generation tools wait up to `wait_seconds`, which defaults to 50 seconds. This avoids resubmitting work when an MCP client reaches its timeout.

Related MCP servers

All in Developer Tools All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewApimodels AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/apimodels?style=directory)](https://allmcps.com/mcp/apimodels)
HTML Embed
<a href="https://allmcps.com/mcp/apimodels"><img src="https://allmcps.com/api/badge/apimodels?style=directory" alt="Apimodels on AllMCPs" /></a>