Lanne Video MCP
& CLI for Claude, ChatGPT & any AI

Generate text-to-video and image-to-video from your prompts in Claude, ChatGPT, Claude Code, Cursor and any MCP client.

New to Claude or ChatGPT video? Claude AI Video Generator guide · Opus 2026 guide · ChatGPT 2026 guide · ChatGPT vs Claude

Copy the Lanne Video connector URL

You’ll paste this URL into your agent in the next step.

https://lannevideo.com/mcp

Go to Claude → Customize

In Claude desktop or claude.ai, open Customize → Connectors. Name it Lanne Video and paste the URL.

Open Claude Customize

Connect, sign in and start

Start creating

Using Claude Code? Prefer the CLI config. Full MCP docs

How we compare

Clip wrappers, finished pipelines, and a multi-model studio MCP that uses the same credits as the Lanne Video studio.

Server	Best for	Output	Pricing note
Higgsfield	One cinematic shot, model zoo	Clip wrapper	Check vendor
AITuber	Publish-ready narrated Shorts	Finished pipeline	Check vendor
fal	Model shopping, pay-per-run	Clip / pay-per-run	Check vendor
Lanne Video	Multi-model studio in Claude & ChatGPT	T2V / I2V (Seedance 2.5, MiniMax H3)	MCP uses your studio plan credits (no API plan needed).

Full write-up: Best MCP servers for AI video (2026). Same MCP for ChatGPT & Claude? Read ChatGPT vs Claude (MCP, 2026).

Connect in 3 steps →

How does MCP work?

Your agent calls Lanne Video tools over Streamable HTTP: generate from text or an image, check status, list videos, and read your credit balance.

You

Create a 9:16 UGC product review: a creator unboxes a skincare serum in a bright bathroom, handheld, captions, natural lighting.

Lanne Video MCP

generate_text_to_video · seedance2 · 9:16 · 8s · audio

Video ready. 9:16 · 8s · Seedance 2.5

Create with Lanne Video in your agent

Give Claude, ChatGPT or Claude Code the same models you use in the studio

Every creative model, inside your agent

The agent picks the right model, or you name it in the prompt. Model deep-dives: MiniMax H3 Video (/minimax-h3-video) for 2K + stereo, and Seedance 2.5 (/seedance-25) for fast social / UGC. Same engines this MCP exposes. Same MCP for ChatGPT & Claude? Read ChatGPT vs Claude for AI Video (MCP, 2026) (/blog/chatgpt-vs-claude-ai-video-mcp-2026) - same Seedance 2.5 + MiniMax H3 backends; agent UX differs.

DEFAULT

Seedance 2.5 Fast

4–15s, audio, 7 aspect ratios. Best for social and UGC.

HIGH QUALITY

MiniMax H3

4–15s, 768P or 2K, 6 aspect ratios. Cinematic and architecture.

Got any questions left?

How does Lanne Video connect to AI agents?

Lanne Video uses MCP (Model Context Protocol), an open standard that gives agents access to external tools. Add the connector URL https://lannevideo.com/mcp, sign in, and your agent can generate videos from text or images, list past generations, and read your credit balance.

Which agents are supported?

Claude, ChatGPT, Claude Code, Cursor, OpenClaw, Hermes Agent, and any client that speaks MCP over Streamable HTTP.

Do I need an API key?

No REST API key is required. Claude-style clients authenticate with OAuth (sign in to your Lanne Video account). CLI clients use a personal MCP token from Dashboard → MCP.

How does pricing work?

MCP uses the same credits as the Lanne Video studio. Each generation deducts credits from your active plan.

What can I create?

Text-to-video and image-to-video clips: Seedance 2.5 (4–15s with audio) and MiniMax H3 (4–15s, 768P or 2K). Vertical 9:16 or landscape 16:9.

How long does generation take?

Most clips are ready in under a minute and the agent returns the video URL directly. Longer renders (FLUX 3 at 1080p, 20-second shots) keep going in the background: the agent checks back and shares the link when it is done.

Can I use this with my own agent?

Yes. Point any MCP-compatible client at https://lannevideo.com/mcp and send a Bearer token, or complete the OAuth sign-in flow.

Lanne Video MCP & CLI for Claude, ChatGPT & any AI

Connect Lanne Video MCP in three steps