NewsAdGraam raises ₹3 crore in seed funding
AI Video & Image NewsConfirmed

Vidu MCP and CLI Are Live: Agent Access to Vidu

AdGraam News Desk5 min read
A terminal and a robotic arm passing film frames to a floating video reel in a dim studio

The short answer

ShengShu (Vidu) announced on November 17, 2026 that Vidu MCP and CLI are officially live, so AI agents and terminal environments can call Vidu directly. The announcement post is short and does not list tools, pricing or supported clients. Pricing, setup steps and supported models are not stated in it.

On this page (11 sections)

What are Vidu MCP and CLI?

Vidu MCP and CLI are new access points for ShengShu's Vidu video generator, announced as officially live on November 17, 2026. MCP connects Vidu to AI agents, and the CLI brings it into terminal environments, so video generation can sit inside an existing agent workflow.

Two cables connecting to a central glowing video generator blockTwo cables connecting to a central glowing video generator block

The announcement came from the official Vidu account on X. It says: "From idea to video, your AI Agent can now handle the workflow," and that Vidu is "now accessible directly through your favorite AI Agents and terminal environments, no constant switching between tools." The post also says Vidu MCP can turn a simple prompt into video, but the text is cut off before the details.

Vidu MCP and CLI release date and availability

Vidu MCP and CLI were announced as live on November 17, 2026, according to ShengShu's official post. That makes this a maker-confirmed launch, not a leak or a teaser.

The post does not name supported agent clients, install commands, authentication, regional limits or whether access needs a paid plan. Those details are not announced yet. Check Vidu's developer documentation at platform.vidu.com before building on it.

Vidu MCP and CLI key features and specs

The confirmed feature is the access layer: prompt-to-video through an agent via MCP, plus a command line option for terminal workflows. Model coverage, output limits and pricing for these interfaces are not stated in the post.

For context, here is what is known about the Vidu models an agent might reach, from third-party reporting:

ModelResolutionDurationAudioNotes
Vidu Q4 Preview (released Oct 7)540p, 720p, 1080p, 2K, 4KImage-to-Video 3 to 16s; Reference-to-Video 1 to 16sNative, on by defaultNo text-to-video; up to 15 reference images and 3 reference audio clips
Vidu Q4 / Q3Up to 1080pNot announcedYesPer ScreenWeaver's description
Vidu S2-Avatar720pReal-time streamConversational25 to 42 fps reported by the official repository
Vidu MCP and CLINot announcedNot announcedNot announcedWhich models it exposes is unconfirmed

The Q4 Preview details come from an OrcaRouter comparison. It notes the endpoints are img2video and reference2video, and neither accepts a prompt alone. If MCP exposes the same endpoints, an agent would need to supply an image or references, not just a sentence. That is an inference, not something ShengShu has said.

Vidu MCP and CLI pricing and free access

ShengShu has not announced separate pricing for MCP or CLI. Usage most likely draws on the same credits as Vidu's existing plans, but that is not confirmed.

Existing rates give a rough floor. A Higgsfield pricing comparison puts a 5 second Vidu Q4 clip at 540p at about $0.13 on the $10 per month Standard plan, checked November 7, 2026. For S2, aiidelist estimates API costs from $0.30 per minute, and says official documentation lists 2,000 free credits for eligible new users, subject to a cap.

How to use Vidu MCP and CLI for AI video ads

For performance teams, the useful case is batch work: an agent generates hook variants, b-roll or product motion clips from a brief, and a CLI lets a script loop over a spreadsheet of SKUs. Neither has been shown in action yet, so treat this as a plan to test.

A laptop assembling abstract storyboard frames into a vertical video stripA laptop assembling abstract storyboard frames into a vertical video strip

Three practical points:

  • Q4 Preview accepts up to 15 reference images, which gives you a way to feed several product shots into one clip. Whether the output keeps the product consistent has not been shown.
  • Native audio on Q4 could help with quick UGC-style tests, though lip sync and voice quality have not been shown here.
  • Return URLs for Q4 creations stay valid for 24 hours, so an automated job should download outputs right away.

If you want to compare Vidu against other video models before committing to one, run the same brief in the AdGraam Playground. You can also chain image, video and upscale steps in Flow Builder, or turn a product URL into scripted video with Video Agent. For hook and hold benchmarks to judge the output, see hook rate and hold rate.

Vidu MCP and CLI vs other ways to access Vidu

Vidu was previously reachable through its web app, phone apps and API. MCP and CLI add agent and terminal routes on top of that, according to the announcement.

ScreenWeaver describes Vidu as prompt or reference first, with Vidu Agent able to take one sentence to a finished piece in conversation. The new MCP route moves that idea into third-party agents. Script-first tools take the opposite approach, starting from a screenplay and storyboard.

For related Vidu coverage, see our pieces on Vidu S2 Avatar and Vidu S2 Editing.

What to watch next

Watch for the full documentation: supported clients, the tool list, credit costs per call and whether Q4 and S2 are both reachable. Also watch whether the final Q4 release adds text-to-video, since the preview lacks it.

Until ShengShu publishes those details, the confirmed fact is narrow: Vidu is live in agents and terminals as of November 17, 2026.

From the timeline

What people are posting

🚀 Vidu MCP & CLI are officially live! From idea to video, your AI Agent can now handle the workflow. Vidu is now accessible directly through your favorite AI Agents and terminal environments — no constant switching between tools. ✨ With Vidu MCP, turn a simple prompt into https://t.co/M1Ynmq6mPj
@ViduAI_official2 likesView on X

Key takeaways

  • ShengShu (Vidu) said on November 17, 2026 that Vidu MCP and CLI are officially live.
  • The launch lets AI agents and terminal environments reach Vidu without switching between tools.
  • The announcement post does not list supported clients, commands, rate limits or separate pricing for MCP and CLI.
  • Vidu Q4 Preview, released October 7, offers up to 4K output and native audio, but only Image-to-Video and Reference-to-Video modes, according to a third-party review.
  • Vidu has been listed at about $0.13 for a 5 second Vidu Q4 clip at 540p on its Standard plan, according to a Higgsfield comparison.
Run today's top models

Compare 200+ AI video and image models side by side

Run Kling, Veo, Sora, Seedance, Flux, Ideogram and Recraft in one Playground, or turn any product URL into ready-to-launch ads.

Questions, answered

Frequently asked

What are Vidu MCP and CLI?

They are two new ways to reach Vidu video generation from outside its web app. MCP connects Vidu to AI agents, and the CLI brings it into terminal environments, according to ShengShu's announcement.

When did Vidu MCP and CLI launch?

ShengShu announced them as officially live on November 17, 2026, through the official Vidu X account.

How much do Vidu MCP and CLI cost?

Separate pricing for MCP and CLI has not been announced in the post. Vidu's standard plan pricing is the closest reference: about $0.13 for a 5 second Q4 clip at 540p, per a Higgsfield comparison.

Which Vidu models can agents call through MCP?

The announcement does not say. Vidu's current lineup includes the Q-series video models and the S2 real-time models, but which ones MCP and CLI expose is not confirmed.

Can marketers use Vidu through an agent for ads?

The announced goal is going from idea to video inside an agent workflow, which fits hooks and b-roll. Commercial terms for agent usage have not been published in the post.

Where this comes from

Sources

  1. 12 Best Affordable AI Video Generators in 2026 (Pricing ...higgsfield.ai
  2. Vidu AI Alternative for Script-First AI Filmsscreenweaver.ai
  3. Vidu S2 - Real-Time Avatars, Live Editing, and API Costsaiidelist.com
  4. Kling 4.0 vs Vidu Q4 Preview: Text-First or Image-Firstorcarouter.ai
AI Video & Image Newsvidu mcpvidu clividu ai agentshengshu viduai video agent workflowvidu q4mcp video generation
Keep reading

More AI model news & creative guides

AI Video & Image News

Vidu S2 Editing: Live Stream Character & Outfit Swaps

Vidu S2 Editing applies character, outfit and style changes to live video. Get the spec table, API cost estimates, ad workflow breakdown and access details.

Nov 16, 2026 · 5 min
AI Video & Image News

Seedance MCP at $0.0918/Sec: YouArt's Pricing Claim

YouArt says its Seedance MCP costs $0.0918/sec pay as you go. Get the price claim vs BytePlus list rates, a spec table, ad workflow notes and access details.

Nov 17, 2026 · 4 min
AI Video & Image News

CapCut Web Ultra: 7 Days Unlimited Seedream Images

CapCut Web Ultra reportedly offers 7 days of unlimited Seedream 5.5 and GPT Image 2.5 generation for Nov 13-19 buyers. Get the spec table, ad workflow tips and access.

Nov 16, 2026 · 4 min
AI Ad Creation

Best AI Ad Generators for D2C Brands in 2026

Ten AI ad generators compared for D2C and e-commerce brands: what each makes, real starting prices, free options, and which suits Indian brands paying in rupees.

Nov 7, 2026 · 11 min
Meta Ads for Indian D2C

Break Even ROAS for COD Orders: How RTO Changes the Math

How to work out break even ROAS when a share of COD orders comes back as RTO, with worked rupee examples, a prepaid and COD mix, and why Ads Manager ROAS overstates COD.

Nov 7, 2026 · 7 min
Meta Ads for Indian D2C

What Indian D2C Brands Run on Meta: 2,790 Ads Analyzed

We studied 2,790 Facebook and Instagram ads from 14 Indian brands and their 747 longest-running ads: formats, angles, offers, languages and CTAs that last.

Nov 7, 2026 · 7 min

Your move

See your product as a winning ad

Turn a product URL into static ads and UGC-style videos, rebuild the competitor ads that already work, and read your Meta data to know what to make next.

AdGraam News DeskAI video and image news

The AdGraam News Desk (part of AdGraam, also searched as Adgram or Adgram AI) tracks AI image and video model launches, leaks and rumors from labs, leakers and creators, and checks each claim against its sources. Unconfirmed stories say so, and each page is updated as its story develops.

LinkedIn