vlogme.ai

Feature map

Everything VlogMe can do

A guided tour of every page — what it's for, how to use it in three steps, and the standout tricks that make it worth your time.

Automation & growth

The parts most people miss

News → Reel → Post

Trends drafts, renders and schedules to your socials on autopilot.

Self-hosted engines

Our own avatar and 9:16 video clusters — faster, cheaper, no black box.

Telegram Mini App

Sign in with your Telegram account and pay with Stars — no forms.

API + MCP for agents

Drive VlogMe from your backend or straight from Claude / Cursor / Codex.

The Dashboard is the map of your account: projects in progress, finished videos, credit balance, and quick launchers into Create, the Studios, the Lab and Social. It's the first thing you see after signing in and the fastest way to jump back into whatever you were doing.

How to use it

  1. 1Sign in — you land here by default.
  2. 2Pick a recent project to keep editing, or hit a Studio tile to start something new.
  3. 3Keep an eye on your credit meter and plan renewal in the corner.

Everything inside

  • Live grid of recent projects and finished renders
  • One-click into Create, Video/Image/Audio Studios and Lab
  • Credit balance, plan and renewal date always visible
  • First-render celebration so wins never go unnoticed
  • Brand DNA card — your voice, style and audience in one place

Create is where your idea becomes a vertical talking-avatar reel. Drop in a portrait, type a wish or paste a script, and a live timeline builds itself card by card. The AI understands photos, voice, motion prompts and even reordering — you can talk to it like a director, or take the wheel yourself.

How to use it

  1. 1Upload one portrait (or skip and start blank).
  2. 2Tell the AI what the video is about — text, voice, or reference images all work.
  3. 3Tweak the timeline, hit Create video, wait a few minutes.

Everything inside

  • Multimodal chat — text, images and voice, all in one thread
  • Top-to-bottom timeline with numbered steps and drag-to-reorder
  • Script Grammar v2 — overlay {@imageX ...}, insert @imageX {}, continue {}
  • Auto voice pick from the portrait's apparent gender & vibe
  • AI can generate its own images and audio when a scene needs them
  • Undo history, live cost estimate and instant Script ↔ Timeline sync
  • Examples on empty state so you never stare at a blank page
  • Chat can also reorder, replace and delete timeline cards for you

Highlights

Director mode chat

Say things like 'swap scene 2 and 3, make the ending in the park' — the AI edits the timeline in place.

Script Grammar v2

Three tiny symbols cover overlay, insert and continue — the AI writes valid grammar and never breaks continuity.

Voice from photo

The right voice is auto-selected from the portrait, so the first playback already sounds like the character.

Video Studio is a curated console for the world's best generative video models. Every task type is one click away and you always see the price before you spend a credit. Perfect for standalone clips or for feeding assets back into a Create project.

How to use it

  1. 1Pick a task: text-to-video, image-to-video, first/last frame, or talking avatar.
  2. 2Pick a model — Veo, Kling, Luma, PixVerse, MiniMax, Hunyuan and more.
  3. 3Click the preview zones to add inputs from Library or upload, then Generate.

Everything inside

  • Task types: T2V, I2V, first & last frame, talking avatar
  • Real model roster with their real names — no black-box guessing
  • Responsive 1/2/3-column layout across phone, tablet and desktop
  • Click-anywhere-to-pick inputs, not just the tiny button
  • Live credit estimate and duration snapping (5s / 9s where required)
  • Human-readable error messages instead of raw provider codes

Highlights

Self-hosted VlogMe Video

9:16 clips up to 8s run on our own Hunyuan cluster — faster and cheaper than routing through third parties.

Talking Avatar

Drop a portrait + a line of speech and get a lip-synced clip — production-ready, not a demo toy.

Image Studio pairs a strong prompt-to-image engine with a full Photo Editor. Generate characters, products and scenes, then pan, zoom, draw and filter them on any device — including iPhone, where multi-touch gestures actually feel right.

How to use it

  1. 1Describe what you want, or upload an existing photo.
  2. 2Open the Photo Editor to adjust, filter, draw or swap the background.
  3. 3Save changes, or 'Save as new' to keep the original untouched.

Everything inside

  • Portrait, product and scene generation from plain text
  • Photo Editor with two-finger pan and pinch-to-zoom
  • Draw, erase, filters and adjustments — with iOS-safe fallbacks
  • Background swap and reframe without leaving the browser
  • Every edit stays in your Library, deduplicated by content hash

Audio Studio covers the full soundtrack: text-to-speech with cloned or curated voices, full songs (yes, actually sung), and one-off sound effects. Anything you generate lands in the Library and can be dragged into any project or video.

How to use it

  1. 1Pick the mode: speech, music/singing, or SFX.
  2. 2Type the line or the song idea, choose a voice or style.
  3. 3Preview, tweak, save — then reuse it anywhere on the site.

Everything inside

  • Text-to-speech with a big library of natural voices
  • ElevenLabs singing — real vocals, not just narration
  • Music generation for hooks, intros and full tracks
  • Sound effects and ambience from a short prompt
  • Accurate duration detection so credit costs are always correct

The Lab is your personal media library plus a live view of anything currently rendering. Content-hash deduplication means you'll never accidentally store the same photo twice, and expired signed URLs re-sign themselves in the background so you never see broken thumbnails.

How to use it

  1. 1Browse or search — filter by type, project or date.
  2. 2Click any tile to open the full-screen viewer with download and edit.
  3. 3Reuse an asset by picking it from any 'Pick from Library' modal.

Everything inside

  • Grid library with fast filters and search
  • Live 'Rendering now' section with progress overlays
  • SHA-256 dedup — same file uploaded twice = one row
  • Click-to-play (no hover autoplay), download that really downloads
  • iOS-safe download via the native Web Share sheet
  • Unified LibraryPicker with an Upload tab, available everywhere

Social turns finished videos into scheduled posts across Instagram, TikTok and YouTube. Ad-hoc publishing, calendar scheduling with your local time zone, analytics and an AI performance coach are all built in.

How to use it

  1. 1Connect your Instagram, TikTok and YouTube accounts.
  2. 2Pick a finished video, write the caption, choose a time (or 'Post now').
  3. 3Track results and let the coach suggest what to try next.

Everything inside

  • Multi-account connection with popup-blocker recovery
  • Ad-hoc publish or full calendar scheduling with time zones
  • Post history, published-platforms badges and analytics
  • Smart schedule suggestions based on your audience
  • Performance Coach — actionable tips per post
  • Reply-video and Reshoot-with-fixes shortcuts from the analytics view

Tutorials are interactive walkthroughs — not videos to watch. They highlight the buttons, prefill the inputs and even trigger a demo render so you finish the lesson with something to show.

How to use it

  1. 1Pick a tutorial — each covers one workflow end-to-end.
  2. 2Follow the on-screen highlights, tap Next between steps.
  3. 3Ship a real render at the end and jump into the next one.

Everything inside

  • Hands-on, in-app step-by-step for every core workflow
  • Demo render dialog so you can see the pipeline in action
  • Next-tutorial prompt keeps momentum after each win

Settings covers the usual — profile, plan, credit history, notifications — plus a full developer suite: personal API tokens and a native MCP (Streamable HTTP) endpoint so Claude, Cursor, Codex and ChatGPT-with-MCP can drive VlogMe directly.

How to use it

  1. 1Set your profile, default language and time zone.
  2. 2Manage your plan and top-up credits (Stars payments if you're in Telegram).
  3. 3Developers: mint an API token or hook the MCP endpoint into your agent.

Everything inside

  • Plan management, upgrade/downgrade and credit history
  • Telegram account linking with recurring subscriptions
  • Personal API tokens for backend integrations
  • MCP Streamable HTTP endpoint for AI agents
  • Full docs: REST + MCP, worked examples in the Manual

Highlights

Public API

Same engine and pricing as the app — ~1 credit per second of output, auto-refunded on failure.

MCP for agents

One endpoint, and Claude / Cursor / Codex can render, edit and publish videos on your behalf.

Ready to try it?

Start with a portrait and a sentence — you can be watching a finished clip in minutes.