Rishikesh from ngram

Anish Muppalaneni

Learn the workspace, create a first video, and give ngram useful instructions and sources.

Start here

ngram is an AI video workspace for turning a prompt, source material, or recording into a script, storyboard, narrated video, and editable project.

What is ngram?

Describe the outcome, add useful source material, choose how much control you want, and let ngram build the first complete version.

Create your first video

Use chat and the unified storyboard workspace together to review progress, inspect scenes, revise the video, edit, download, and share.

Understand the agentic workspace

Give ngram the goal, audience, source of truth, and creative constraints that matter—and leave room for the agent to plan the rest.

Write a better video prompt

Start with a fill-in brief for a demo, explainer, update, social clip, training video, or presenter-led story—then replace every bracket with your own approved information.

Use prompt recipes for common videos

First separate a technical failure from an expectation mismatch, then repair the brief or request the smallest useful revision without discarding good work.

Fix a prompt that missed the brief

Ground a video in public URLs or files you are allowed to use, then tell ngram which facts and visual assets should drive the result.

Use sources and attachments

Get step-by-step guides, tutorials, and tips to create stunning product demo videos with ngram.com. Find answers to all your questions in one place.

Community forum

Ngram.com

Choose the approval flow, video type, models, style, identity, duration, and research settings.

Create a video

YOLO keeps moving through to a video. PRO pauses after planning so you can approve the storyboard before visual generation begins.

Choose between YOLO and PRO

Pick the production path that matches the source material and the result you want, rather than choosing by the underlying rendering technology.

Choose a video type

Use the current Studio selectors to balance quality, speed, input support, and credits across text, image, and video generation.

Choose AI models

Let Auto pick a suitable look or choose a visible style preset when consistency with a format or brand direction matters.

Choose a visual style

Use the right reusable layer: a template for a starting structure, a style for visual treatment, a theme for your saved creative system, and Brand Kit for approved identity assets.

Choose between templates, themes, styles, and Brand Kit

Preview a narration voice, choose the language and avatar behavior, or use custom identity assets when your plan includes them.

Choose a voice and avatar

Choose a target length and orientation that fit the message, destination, and limits shown for your current plan.

Set duration and aspect ratio

Turn on Deep Research when image-heavy source material needs a more intensive visual-analysis pass than the normal workflow.

Use Deep Research

Correct the story, claims, narration, and scene plan before asking ngram to spend more time and credits on visual generation.

Review scripts and storyboards

Revise with chat, edit scenes directly, use the timeline, adjust sound, and export the right version.

Edit and export

Start with chat for intent, use frame and scene targeting for precision, then switch to direct editing only where it saves time.

Revise a generated video

Use the inline Edit workspace for direct scene changes while keeping the work connected to the original conversation.

Edit inside an agentic conversation

On Ultimate, add a quiz that pauses playback or a timed action button that opens a link, then publish the draft and review viewer responses.

Add quizzes and action buttons to a video

Create a separate editor copy when you need detailed manual control over the canvas, media, pages, effects, and timeline.

Use the full video editor

Navigate pages and tracks, move the playhead, adjust supported items, and review timing in both inline and full-editor timelines.

Work with the timeline

Use the quick Edit video menu or the editor panels to control the audio and caption layers of a finished video.

Adjust music, voiceover, captions, and sound

Download the current finished video as an MP4 at a resolution your plan supports, and verify that the file matches the latest saved changes.

Download and export a video

Manage projects, team access, brand systems, and custom voice or presenter assets.

Organize and collaborate

Use Recents for quick return and Projects for a complete, filterable view of agent chats and one-shot tool outputs.

Manage projects, conversations, and Tool results

Preserve good work before a major change, understand which history controls are reversible, and avoid editing the wrong copy.

Use versions, forks, undo, and editor copies

Create an Ultimate team, invite members, assign roles, share work and assets, and manage sponsored plans and ownership.

Use teams and collaboration

Store approved logos, mascots, colors, typography, and intro or outro assets so videos can begin from the same brand source.

Set up a Brand Kit

Add a permitted voice sample in Settings, then select the resulting voice from supported narration workflows.

Create and manage a custom voice

Upload a permitted presenter image in Face Settings, then use that face in compatible avatar and talking-head video workflows.

Create and manage a custom face

Publish a public watch link, embed or download a video, and connect supported social channels.

Share

Control the public watch link, copy an embed, download the current render, publish to connected channels, and republish after revisions.

Share, embed, publish, and update a video

Connect supported accounts from Integrations, then publish a finished video from Share without downloading and re-uploading it manually.

Connect and publish to social channels

Use focused creation, image, clip, recording, and asset-repurposing workflows.

Tools and apps

Find a focused generator or editing action, understand whether it opens a dedicated workflow or the agentic composer, and locate the result afterward.

Use the Tools library

Use a dedicated Tool when you need one clear change—such as eye contact, background cleanup, transcription, translation, voice dubbing, or media conversion—without rebuilding a full video.

Edit existing media with focused AI Tools

Upload a permitted talking-head video and create a new version in which the speaker appears to address the lens more directly.

Correct off-camera gaze with Eye Contact AI

Choose the focused workflow that matches the output: a localized video, lip-synced translation, replacement voice track, or searchable transcript.

Translate, dub, or transcribe existing media

Generate brand-aligned business images from a prompt or template, attach references, choose a platform shape and model, then download the result.

Create images in Image Lab

Generate a focused clip from a prompt, one image, start and end frames, or a set of references with a compatible video model.

Create short AI video clips in Clip Studio

Capture a product walkthrough in the browser, review the recording, then add it as source video for editing or AI-assisted polishing.

Record your screen and create a polished screencast

Turn a finished paid-plan project into reusable per-scene clips, audio, captions, images, text, and project assets without rebuilding them one by one.

Download a Repurpose Kit

Understand plans and credits, inspect usage, configure your account, and use current reward programs.

Account and billing

Compare current monthly or annual credit allocations, export quality, identity assets, team access, and top-up pricing in Billing & Pricing.

Understand plans and pricing

Credits pay for AI generation and processing. The final cost depends on the operation, plan, selected model, generated duration, and current product configuration.

Understand credits

See available and used credits, inspect human-readable activity, open the related project, and review the last 30 days of usage.

Review your usage history

Review profile and sign-in information, configure job behavior and notifications, move between settings areas, or schedule account deletion.

Manage your account and preferences

Open Earn credits to see the actions or referral program currently enabled for your account, and treat the live terms as authoritative.

Earn credits and check reward programs

Recover durable generation and export jobs and give Support the details needed to help.

Troubleshooting

Open Help when you need a guide; when Support is visible, use it for a project, account, payment, or product control that needs direct assistance.

Use Help and contact Support

Long-running work is tied to its conversation. Leave and return safely, identify a true failure, retry the smallest useful step, and share useful details with Support.

Understand generation status, retries, and recovery

Create self-service API keys and connect the API, MCP, Zapier, n8n, and other available integrations.

Automations and developer

Connect publishing channels, install the Chrome extension, or use available automation surfaces while respecting current authentication gates.

Understand integrations and automation access

Create a bearer key in Settings, start a render, and collect the finished MP4 from any language that speaks HTTP.

Get started with the ngram API

Learn how every signed-in ngram user can create, copy, rotate, and revoke an API key for the REST API, MCP, Zapier, and n8n workflows.

Create and manage your ngram API key

ngram is a native Model Context Protocol server, so Claude, eligible ChatGPT web workspaces and authenticated MCP-aware clients can create videos, check status and read credits as tool calls.

Connect ngram over MCP

Connect ngram to thousands of apps without writing code. Trigger a render when a row is added, a form is submitted or a deal closes — then send the finished video wherever your team already works.

Automate ngram with Zapier

The ngram community node, verified by n8n, lets n8n workflows create videos and react to render events. Install it from the Nodes panel on self-hosted n8n or n8n Cloud, add an API key from Settings, and drop the nodes into a workflow.