Tangra AI Video Generator

Create top-quality AI videos instantly - no filming, no editing, no costly tools.

Overview

Tangra AI Video Generator turns a text prompt or a still image into a finished video clip using a large catalog of state-of-the-art AI video models (Veo, Sora, Kling, Seedance, Vidu, Pixverse, Minimax, Wan and more). You can generate from text, animate your own images (image-to-video and multi-image-to-video), pick the duration, aspect ratio and resolution, then assemble multiple clips, images, AI music and AI narration on a timeline-based editor and export an MP4 or WebM. A Story Wizard can turn a written story into scenes, images, narration and music in one pass.

Who it’s for

Content creators and social media managers who need short-form video (TikTok, Reels, Shorts, YouTube)
Marketers and small businesses producing ads and promos without a film crew
Storytellers and educators turning scripts or stories into narrated videos
Designers and artists who want to animate their own images
Anyone who wants professional-looking video without editing software or filming

Key features

Text-to-video generation

Type a prompt and an AI model generates a video clip. Supported by text-to-video models such as Veo 3.1 / Veo 3.1 Fast, Sora 2 / Sora 2 Pro and Reel 1.1.

Image-to-video and multi-image-to-video

Upload one or more images and animate them. Models like Kling 2.1 (Standard/Pro/Master), Seedance 1.5 Pro, Vidu, Pixverse 5.5, Minimax 2 and Wan turn stills into motion; some accept first/last frame images or multiple reference images (Pro 2 supports up to 9 reference images).

Large multi-model catalog

A single picker exposes many AI video engines grouped by use case (Bring Your Image to Live, Video from text, New) with per-model credit cost, duration limits and aspect-ratio options shown inline.

Adjustable duration, aspect ratio and resolution

A length slider is bounded by each model's min/max seconds and step. Aspect ratios range from 9:16, 1:1, 4:3, 16:9, 21:9 up to 4K (3840x2160 / 2880x2880 / 2160x3840) depending on the model.

Timeline video editor

Drag generated clips, images, AI music and AI voice narration onto a multi-track timeline. Trim, reposition and layer items, add transitions (fade, slide, wipe, zoom, rotate) and per-clip effects (beat pulse, color modulation, camera pan/zoom presets).

Text and rich-text overlays

Add titles, captions and styled text layers directly on the canvas, with positioning, rotation and opacity controls.

AI music and narration

Generate background music and AI voice-over from within the editor (music gallery, voice gallery, music wizard) and beat-sync visual effects to the audio.

Multiple export paths

Export in the browser with Quick (fast MediaRecorder) or Advanced (hardware-accelerated WebCodecs with codec choice), or offload a full-fidelity Server Render. Output as MP4 or WebM up to 4K.

Story Wizard

Turn a written story into a complete video — Tangra generates scenes, images, narration and music in one guided flow, with a Simple and an Advanced mode.

How to use it — step by step

Open the AI Video generator
Go to tangra.link/video (landing) and launch the app at /video/app. The prompt box and model picker are the main controls.
Choose how to generate
Pick a mode via the model filter: Text to Video, Image to Video, or Multi-image to Video. This filters the model list to engines that support that mode.
Select an AI model
Open the model selector and choose a model (e.g. Veo 3.1, Sora 2, Kling 2.1, Seedance, Pixverse, Wan), or leave it on 'Select AI' to let Tangra pick. Each model shows its credit cost, length range and aspect ratios.
Add a prompt (and images if needed)
Type your description in the prompt box. For image-to-video models, click the + button to add the required reference image(s); some models accept first and last frame or multiple references.
Set duration and aspect ratio / resolution
Use the length slider (limited to the model's min/max and step) to set seconds, and choose an aspect ratio / size from the model's available options (portrait, square, landscape, up to 4K).
Generate the clip
Click Generate. Tangra renders the video; the result appears in your gallery of recent generations and can be replayed.
Open the video editor (optional)
Go to /video/editor to drag your clips, images, AI music and AI narration onto the timeline. Trim, reorder, add transitions, text overlays and effects, and preview with Play.
Export your video
Click Export and choose Quick, Advanced (WebCodecs) or Server Render. Pick resolution, quality and format, then download the finished MP4 or WebM.

See it in action

Screens you’ll work with inside Video Generator:

Video landing page — Marketing hero for the AI Video Generator showing the value proposition and a call to action to start creating. — Fig 1. Video landing page

Video generator app — The working app: prompt box, generation mode filter (Text/Image/Multi-image to Video), model selector, length slider, aspect ratio/size picker, Generate button and gallery of recent generations. — Fig 2. Video generator app

Model selector open — The model picker expanded, showing grouped AI models (Bring Your Image to Live, Video from text, New) with names, credit costs and thumbnails. — Fig 3. Model selector open

Timeline video editor — The multi-track timeline editor with a clip on the canvas, tracks for video/image/music/voice, the tool panel (text, music, voice, AI edit), and the playhead/Play controls. — Fig 4. Timeline video editor

Export Video modal — The export dialog showing Quick / Advanced / Server Render modes, resolution dropdown with detected aspect ratio, quality presets, format/codec options and estimated time/size. — Fig 5. Export Video modal

Story Wizard — The Story Wizard flow that turns a written story into scenes, images, narration and music (Simple and Advanced modes). — Fig 6. Story Wizard

Video pricing — Plans and credit packs for generating AI videos (pay-as-you-go and monthly plans). — Fig 7. Video pricing

Settings & controls

Every control in Video Generator, what it does and the options available:

Control	What it does	Options
Generation mode filter	Filters the model list by capability.	Text to Video, Image to Video, Multi-image to Video
AI model selector	Choose the AI video engine; shows credit cost, duration range and aspect ratios per model.	Select AI (auto), Veo 3.1, Veo 3.1 Fast, Sora 2, Sora 2 Pro, Reel 1.1, Kling 2.1 Standard/Pro/Master, Kling O1, Seedance 1.5 Pro, Fast, Pro 2, 4K, Vidu 2, Vidu Q2 Pro, Pixverse 5.5, Minimax 2, Wan 2.2, Wan 2.6
Prompt box	Text description of the video to generate.	Free text
Reference image(s)	Add input images for image-to-video / multi-image models via the + button.	1 to 9 images depending on model (some support first/last frame)
Length (duration) slider	Sets clip length in seconds, bounded by the selected model.	Varies by model, e.g. 4-8s (Veo 3.1), 5-10s (Kling), 3-15s (4K), 6-120s (Reel 1.1)
Aspect ratio / size	Output dimensions and aspect ratio for the generated clip.	9:16, 1:1, 4:3, 16:9, 21:9 and up to 4K, per model
Timeline tracks	Multi-track editor for layering video, images, music and voice.	Video, image, music and voice/narration tracks
Transitions	Transition applied between overlapping clips.	Fade, Slide (left/right/up/down), Wipe (left/right/up/down), Zoom, Rotate
Clip effects	Per-clip motion and color effects, can be beat-synced to audio.	Beat pulse, Color modulation (brightness/contrast/saturation), Camera presets (zoom-in, zoom-out, pan-left, pan-right), Section choreography
Text overlays	On-canvas titles and captions with positioning.	Text and Rich Text tools
Export mode	How the final video is rendered.	Quick (MediaRecorder), Advanced (WebCodecs, hardware-accelerated), Server Render (cloud)
Export resolution	Output resolution, offered per aspect ratio with a recommended pick.	Original, 480p, 720p, 1080p, 1440p/2K, 4K (varies by aspect ratio)
Export quality	Bitrate/quality preset affecting file size.	Low, Medium, Balanced (High), Ultra
Export format / codec	Container and codec for the output file.	Auto, MP4 (H.264/H.265), WebM (VP9); WebCodecs offers HEVC/AV1/H.264; Server Render offers 24/30/60 fps
Server render quality	Sub-mode for cloud render.	Fast (ffmpeg filter_complex) or Full Quality (frame-by-frame with all effects)

AI models

Video Generator lets you pick from a curated lineup of AI models — each shows its credit cost in the app. Available models:

Veo 3.1 (text-to-video, image-to-video, multi-image-to-video)Veo 3.1 FastSora 2 (text-to-video)Sora 2 Pro (text-to-video)Reel 1.1 (text-to-video, up to ~120s)Lite / Kling 2.1 Standard (image-to-video)Pro / Kling 2.1 Pro (multi-image-to-video)Studio / Kling 2.1 Master (image-to-video)Kling O1Seedance 1.5 Pro (image-to-video)Fast (Seedance-based image-to-video)Pro 2 (image-to-video + multi-image, up to 9 reference images)4K (Kling 3.0 4K, up to 15s, image-to-video + multi-image)Vidu 2 (image-to-video)Vidu Q2 Pro (image-to-video)Pixverse 5.5 (image-to-video)Minimax 2 (image-to-video)Wan 2.2 (image-to-video)Wan 2.6 (image-to-video)Select AI (auto-picks a model for you)

Outputs & formats

MP4 (H.264 / H.265 export)
WebM (VP9 export)
Resolutions from 480p / SD up to 4K Ultra HD (3840x2160)
Aspect ratios: 16:9, 9:16, 1:1, 4:5, 4:3, 2:3, 21:9, 2.39:1 cinema
Generated clip lengths from a few seconds up to ~120s depending on model
Frame rates for server render: 24, 30 or 60 fps

Pro tips

Use 'Select AI' if you're unsure which engine fits — Tangra auto-picks a model for your prompt.

Match your reference image's aspect ratio to the model's output size; several models (e.g. Kling Pro, 4K) require the image and video sizes to match.

For social platforms pick the right aspect ratio up front: 9:16 for Reels/TikTok/Shorts, 1:1 for square posts, 16:9 for YouTube.

Fast/cheap models (Veo 3.1 Fast, Lite/Kling Standard, Seedance) are great for drafts; switch to Pro/Studio/4K once you like the result, since credit cost scales with model, resolution and length.

In the editor, enable beat-pulse or color-modulation effects on clips and link them to your AI music track so visuals react to the beat.

For the cleanest final file use Advanced (WebCodecs) export or Server Render at Full Quality; use Quick export for fast social previews. Keep the export window open until it finishes.

Video Generator FAQ

What's the difference between text-to-video and image-to-video?

Text-to-video generates a clip purely from your written prompt (e.g. with Veo 3.1, Sora 2 or Reel 1.1). Image-to-video animates one or more images you provide (e.g. Kling, Seedance, Pixverse, Wan); multi-image-to-video uses several reference or first/last-frame images. The mode filter shows only the models that support each option.

How long can a generated video be?

It depends on the model. Many models produce 4-12 second clips (for example Veo 3.1 is 4-8s, Kling models 5-10s, the 4K model 3-15s), while Reel 1.1 can go up to about 120 seconds. The length slider only lets you pick durations the chosen model supports. You can also stitch multiple clips together in the timeline editor for a longer final video.

Which aspect ratios and resolutions are supported?

Generation supports portrait (9:16), square (1:1), standard (4:3), landscape (16:9), ultra-wide (21:9) and more, with per-model size lists. On export you can choose resolutions from 480p up to 4K Ultra HD (3840x2160), plus cinema ratios like 2.39:1, with a recommended option highlighted for each aspect ratio.

What models can I choose from?

Tangra includes a large catalog: Veo 3.1 and 3.1 Fast, Sora 2 and Sora 2 Pro, Reel 1.1, Kling 2.1 Standard/Pro/Master and Kling O1, Seedance 1.5 Pro, Vidu 2 and Vidu Q2 Pro, Pixverse 5.5, Minimax 2, Wan 2.2 and 2.6, plus Fast, Pro 2 and 4K variants — or 'Select AI' to auto-choose.

Can I edit clips together and add music or narration?

Yes. The timeline editor lets you drag generated clips, images, AI-generated music and AI voice narration onto multiple tracks, trim and reorder them, add transitions and effects, overlay text, and preview before exporting.

What file formats can I export?

You can export MP4 (H.264/H.265) or WebM (VP9). Quick export uses your browser's MediaRecorder for speed, Advanced export uses hardware-accelerated WebCodecs with codec choice (HEVC/AV1/H.264), and Server Render produces a full-fidelity file in the cloud at 24, 30 or 60 fps.

What is the Story Wizard?

The Story Wizard at /video/story turns a written story into a finished video in one guided flow — Tangra generates scenes, images, narration and music together. It offers a Simple mode and an Advanced mode.

How much does it cost?

Generation uses credits, and the cost varies by model, resolution and length (each model shows its credit cost in the picker). Tangra offers both pay-as-you-go and monthly plans, which you can compare on the video pricing page.