Create images and videos from your AI assistant
Add VerseIn to ChatGPT, Claude, Claude Code, Codex or Cursor. Sign in once with your VerseIn account, no API key needed, then ask for images and videos right in the chat.
Connect your assistant
https://mcp.versein.ai/mcpOne click opens Claude with VerseIn already filled in. Once added, it also works in the Claude desktop and mobile apps.
Select Connect to Claude. Claude opens Add custom connector with the name VerseIn and the connector URL already filled in.
Check that the name and URL match this page, then add the connector. Select Connect, sign in to VerseIn and select Allow.
In a chat, turn on VerseIn from the tools menu and ask for an image or video.
Adding it by hand instead? In Claude, open Connectors, select Add custom connector, enter the name VerseIn and paste the connector URL.
https://mcp.versein.ai/mcpUntil VerseIn is listed in the ChatGPT app directory, add it as a custom connector in developer mode.
In ChatGPT, open Settings → Apps & Connectors → Advanced settings and turn on Developer mode.
Back in Apps & Connectors, select Create. Name the connector VerseIn and paste the connector URL.
Select Create, sign in to VerseIn in the window that opens and select Allow.
In a new chat, pick VerseIn from the + menu and describe the image or video you want.
https://mcp.versein.ai/mcpInstall in one click, or add the server to mcp.json yourself.
Select Add to Cursor and confirm the install in Cursor.
Or add it by hand: open Cursor Settings → MCP and add this to mcp.json:
mcp.json{ "mcpServers": { "versein": { "url": "https://mcp.versein.ai/mcp" } } }Next to versein, select Needs login, sign in to VerseIn and select Allow.
https://mcp.versein.ai/mcpOne terminal command adds VerseIn to Claude Code.
Run this command in your terminal:
claude mcp add --transport http versein https://mcp.versein.ai/mcpStart Claude Code, type /mcp, select versein and finish signing in through your browser.
Ask Claude Code for what you need, for example a hero image for your landing page.
https://mcp.versein.ai/mcpAdd VerseIn to Codex through its configuration file.
Add this to ~/.codex/config.toml:
~/.codex/config.toml[mcp_servers.versein] url = "https://mcp.versein.ai/mcp"Sign in and approve the connection:
codex mcp login verseinStart Codex and ask it to generate an image or video.
No API key needed. You sign in to VerseIn in your browser and can disconnect at any time.
Create without leaving the chat
Describe what you want. Your assistant picks the model and settings, brings back the result and keeps working with it.
Start from your own product
Attach a product photo, a sketch or a reference and your assistant builds on it, so every new image and video stays true to the original.
Create inside your project
In Claude Code, Codex or Cursor, your assistant already knows what you're building. Ask for assets where you need them and keep going with the results.
Ask for up to 12 at once
Request a whole set of variations in one message. Each one runs as its own task, so you can compare directions side by side.
Know the cost before it runs
Your assistant checks your balance and what a request will cost before generating anything, and larger requests ask you to confirm first.
The same models as the web studio
Your assistant picks from the image and video models on VerseIn, at the same prices you see on the website. A few models may be unavailable through assistants.
Seedance 2.5
MiniMax H3 Max
Seedance 2
Wan 3.0
Veo 3.1
Seedance 2.0 Fast
Veo 3.1 Fast
Grok Video 1.5
Nano Banana 2.1
GPT Image 2.5 Sunburst- GPT Image 2.5 Flare
Seedream 5.0 Flash
Why connect VerseIn to your assistant
No API keys
You sign in with OAuth in your own browser. Your assistant gets a revocable token and never sees your password.
Disconnect any time
Remove an assistant from Connected apps on your Account page, and its access ends immediately.
Follow every task
Follow progress, list recent work and cancel a job that's still running.
Saved to your library
Every result is also saved to your VerseIn library, ready to keep editing, download or publish.
The same rules as VerseIn
Requests from an assistant go through the same content rules, moderation and credit prices as the website.
Likeness stays in VerseIn
Face swap and other features that use a real person's likeness are not available to assistants.
Edit what you already have
Bring your own images and videos, or build on a result, without starting over.
Relight
Upscale
Edit video
Depth map
Reframe
Illustrative examples. Results depend on your source and settings.
See every tool your assistant can call
VerseIn gives your assistant a focused set of tools. It picks the right one for each request, and every call shows up in the chat.
- Create
- Images and videos from a prompt, a reference image, or a first and last frame.
generate_imagegenerate_video- Edit and enhance
- Upscale, relight, reframe or extend your media, edit a video from a text instruction, and turn a video into a depth map.
upscale_imageupscale_videorelight_imagerelight_videoreframe_videoextend_videoedit_videocreate_depth_map- Your media
- Upload a file from your device or bring in media from a link.
create_uploadimport_media_from_urlpick_media- Tasks
- Follow progress, list recent work and cancel a job that's still running.
get_taskslist_recent_tasksopen_libraryopen_thread_traycancel_task- Models and credits
- Browse the models, check what a request will cost and see your balance.
list_modelsget_modelpick_modelestimate_costget_balancesettings_read
How credits work in your assistant
The connector uses the credits in your VerseIn account. Nothing is sold inside an AI assistant.
- Each generation uses the same credits, at the same prices, as the VerseIn website and apps.
- Your assistant checks the cost before it generates, and larger requests ask you to confirm first.
- Generations started from an assistant are always charged in credits. The Unlimited plan does not apply to them.
- If a generation fails for technical reasons, its credits are returned automatically.
- You get credits with a VerseIn membership or credit pack on the VerseIn website or in the mobile app. Bonus credits, such as the connection bonus, count too.
Connection bonus
When a connection bonus is running, the approval screen shows how many free credits you'll receive and when they expire. It's granted once per VerseIn account, after your first connection completes.
Frequently asked questions
Is it safe? Do I need an API key?
You never need an API key. You sign in to VerseIn in your own browser and approve what the assistant can do; the assistant receives a revocable token, never your password. You can disconnect any assistant at any time from Connected apps on your Account page, and its access ends immediately.
How am I billed?
Each generation uses credits from your VerseIn balance at the same prices as on the website, and your assistant can show you the cost first. The Unlimited plan does not apply to generations started from an assistant. Credits spent on generations you authorized are not refunded, but failed generations are refunded automatically.
Which assistants are supported?
ChatGPT (in developer mode for now), Claude on the web, desktop and mobile, Claude Code, Codex and Cursor. Other clients that support remote MCP servers with OAuth sign-in can use the same URL.
Why can't I use face swap from an assistant?
Features that put a real person's face or likeness into content, such as face swap, character and template swaps and motion transfer, are only available in VerseIn itself, on the web and in the apps, where consent is part of the flow. From an assistant you can generate and edit images and videos from prompts and your own media.
Where do my results go?
Your assistant gets a link to each result, and every result is saved to your VerseIn library, where you can keep editing, download or publish it. The same watermark rules apply as on the website.
Connect once. Create anywhere.
Add VerseIn to the assistant you already use, and turn your next idea into an image or video without leaving the chat.
https://mcp.versein.ai/mcp