> For the complete documentation index, see [llms.txt](https://docs.flora.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.flora.ai/models/video-models.md).

# Video Models

Here are all the video models currently available in FLORA.

## Generation Models

These models generate video from text or image inputs.

<table><thead><tr><th width="220">Model</th><th width="120">Est. time (s)</th><th>Description</th><th>Best for</th></tr></thead><tbody><tr><td><strong>Aurora</strong></td><td>30</td><td>Animate an avatar using an image and audio input.</td><td>Animate an avatar using an image and audio input.</td></tr><tr><td><strong>Boreal</strong></td><td>240</td><td>Product, UGC, and presenter videos with synchronized native audio, steered by an optional image and audio track.</td><td>Product, UGC, and presenter videos with synchronized native audio, steered by an optional image and audio track.</td></tr><tr><td><strong>FLUX 3</strong></td><td>180</td><td>Video with synchronized audio, opening and closing frames, timed keyframes, and clip continuation.</td><td>Video with synchronized audio, opening and closing frames, timed keyframes, and clip continuation.</td></tr><tr><td><strong>FLUX 3 Keyframes</strong></td><td>180</td><td>Video with synchronized audio, opening and closing frames, timed keyframes, and clip continuation.</td><td>Video with synchronized audio, opening and closing frames, timed keyframes, and clip continuation.</td></tr><tr><td><strong>Gemini Omni 1.1 Flash</strong></td><td>180</td><td>Google's Omni video model, with 360p-4k output, explicit first and last frames, and video references for character consistency.</td><td>Google's Omni video model, with 360p-4k output, explicit first and last frames, and video references for character consistency.</td></tr><tr><td><strong>Grok Imagine</strong></td><td>115</td><td>Video generation with audio from xAI.</td><td>Quick generation with xAI's latest image capabilities</td></tr><tr><td><strong>Happy Horse 1.0</strong></td><td>250</td><td>Generate videos with synced audio using customizable parameters.</td><td>Generate videos with synced audio using customizable parameters.</td></tr><tr><td><strong>HeyGen Video 1.0</strong></td><td>240</td><td>Short clips with synchronized audio from a prompt, a start frame, or image, video, and audio references.</td><td>Short clips with synchronized audio from a prompt, a start frame, or image, video, and audio references.</td></tr><tr><td><strong>Kling 2.0 Master</strong></td><td>330</td><td>Reliable motion with good consistency.</td><td>Complex long-form sequences, cinematic pre-viz, style explorations.</td></tr><tr><td><strong>Kling 2.1 Master</strong></td><td>330</td><td>Precise prompt following with dynamic motion.</td><td>TikTok-style transitions, before-and-after promos, precise framing.</td></tr><tr><td><strong>Kling 2.1 Pro</strong></td><td>330</td><td>Precise prompt following with dynamic motion.</td><td>High-quality generation with strong prompt adherence.</td></tr><tr><td><strong>Kling 2.5 Turbo Pro</strong></td><td>250</td><td>Smooth motion at faster generation speeds.</td><td>Fast, high-quality generation with smooth motion.</td></tr><tr><td><strong>Kling 2.6 Pro</strong></td><td>200</td><td>Top-tier visuals with synced voice and sound.</td><td>High-quality videos with synchronized audio, music videos, commercials.</td></tr><tr><td><strong>Kling 3.0 Pro</strong></td><td>270</td><td>Animate between start and end frames.</td><td>Premium video with synchronized audio, cinematic quality.</td></tr><tr><td><strong>Kling 3.0 Pro (Turbo)</strong></td><td>270</td><td>Improved lipsync and multishot generation capabilities.</td><td>Improved lipsync and multishot generation capabilities.</td></tr><tr><td><strong>Kling 3.0 Standard</strong></td><td>220</td><td>Faster and more cost effective than Kling 3.0 Pro.</td><td>Quick high-quality generation at lower cost.</td></tr><tr><td><strong>Kling 3.0 Standard (Turbo)</strong></td><td>270</td><td>Optimized for rapid iteration and high-volume production.</td><td>Optimized for rapid iteration and high-volume production.</td></tr><tr><td><strong>Kling Avatar v2 Pro</strong></td><td>300</td><td>Create avatar videos with realistic humans, animals, cartoons, or stylized characters.</td><td>Avatar-driven storytelling, character videos, and stylized spokesperson content.</td></tr><tr><td><strong>Kling O1</strong></td><td>330</td><td>Precise edits with text, image, or video guidance.</td><td>Detailed image editing with high precision</td></tr><tr><td><strong>Kling O3 Pro</strong></td><td>265</td><td>Animate between start and end frames.</td><td>Maximum quality Kling output, premium productions.</td></tr><tr><td><strong>Kling O3 Standard</strong></td><td>220</td><td>Kling O3 at a lower cost and faster speed.</td><td>Cost-effective O3 quality, faster generation.</td></tr><tr><td><strong>Kling Pro 1.5</strong></td><td>300</td><td>Strong motion with dynamic camera work.</td><td>Dynamic camera movements, action sequences.</td></tr><tr><td><strong>Kling Pro 1.6</strong></td><td>360</td><td>Balanced quality and generation speed.</td><td>Product demos, architectural walk-throughs, realistic lighting.</td></tr><tr><td><strong>LTX-2 Pro</strong></td><td>120</td><td>Fast 4K output with generated audio.</td><td>Quick video with synchronized audio, up to 4K resolution.</td></tr><tr><td><strong>Luma Ray 2</strong></td><td>200</td><td>Realistic physics and natural motion.</td><td>Lifestyle, nature, travel footage, drone-style glides, fashion runways.</td></tr><tr><td><strong>Luma Ray 2 Flash</strong></td><td>60</td><td>3× faster than Ray 2, good quality.</td><td>Quick variant testing, social-media cut-downs, preview edits.</td></tr><tr><td><strong>Marey</strong></td><td>420</td><td>Commercially-licensed training data.</td><td>Brand-safe commercial spots, legal-clear hero sequences.</td></tr><tr><td><strong>MiniMax H3</strong></td><td>300</td><td>MiniMax video with keyframe control and image, video, and audio references.</td><td>MiniMax video with keyframe control and image, video, and audio references.</td></tr><tr><td><strong>MiniMax H3 Max</strong></td><td>30</td><td>fal's post-trained H3: sharper prompt adherence at 768P, generated in seconds.</td><td>fal's post-trained H3: sharper prompt adherence at 768P, generated in seconds.</td></tr><tr><td><strong>MiniMax H3 Max Extend</strong></td><td>60</td><td>Continue a clip: H3 Max adds 5 to 15 seconds of new footage and sound after the source, up to 2K.</td><td>Continue a clip: H3 Max adds 5 to 15 seconds of new footage and sound after the source, up to 2K.</td></tr><tr><td><strong>MiniMax H3 Max Recast</strong></td><td>180</td><td>Recast a clip: H3 Max swaps the people in a video for people from your photos, keeping the motion, camera and sound.</td><td>Recast a clip: H3 Max swaps the people in a video for people from your photos, keeping the motion, camera and sound.</td></tr><tr><td><strong>MiniMax H3 Max Turbo</strong></td><td>20</td><td>fal's faster, cheaper H3 Max: 2x the speed at half the cost, 768P in seconds.</td><td>fal's faster, cheaper H3 Max: 2x the speed at half the cost, 768P in seconds.</td></tr><tr><td><strong>Minimax Hailuo</strong></td><td>240</td><td>Strong motion with dynamic camera work.</td><td>Narrative shorts (30-60s), coherent character arcs, evolving lighting.</td></tr><tr><td><strong>Minimax Hailuo 02 Pro</strong></td><td>420</td><td>Native 1080p with realistic physics.</td><td>Ultrafast generation of vertical reels, bumper ads, A/B variants.</td></tr><tr><td><strong>Minimax Hailuo 2.3 Pro</strong></td><td>300</td><td>1080p native with expressive motion.</td><td>High-resolution dynamic content, expressive animations.</td></tr><tr><td><strong>Pika</strong></td><td>350</td><td>Reliable prompt following, up to 10s clips.</td><td>Avatar-led explainers, talking-head shorts, expressive faces.</td></tr><tr><td><strong>Runway Gen-4.5</strong></td><td>180</td><td>High-fidelity cinematic motion with strong prompt adherence.</td><td>Cinematic output, consistent character and scene quality.</td></tr><tr><td><strong>Seedance 1.0 Pro</strong></td><td>165</td><td>High quality with fixed camera option.</td><td>Premium social ads, stylised title sequences, cinematic game-trailers.</td></tr><tr><td><strong>Seedance 1.5 Pro</strong></td><td>120</td><td>Powerful model with native sound.</td><td>High quality with built-in audio generation.</td></tr><tr><td><strong>Seedance 2.0</strong></td><td>250</td><td>Generate a video by setting its start and end frame, via Enhancor.</td><td>Premium social ads, cinematic sequences with synchronized audio.</td></tr><tr><td><strong>Seedance 2.0 Fast</strong></td><td>175</td><td>Generate a video by setting its start and end frame, via Enhancor.</td><td>Quick iterations on Seedance 2.0 quality content.</td></tr><tr><td><strong>Seedance 2.0 Mini</strong></td><td>250</td><td>Lightweight, cost-effective video generation with audio.</td><td>Lightweight, cost-effective video generation with audio.</td></tr><tr><td><strong>Seedance 2.5</strong></td><td>300</td><td>Single-shot video up to 30 seconds with native audio.</td><td>Single-shot video up to 30 seconds with native audio.</td></tr><tr><td><strong>Sora 2 Pro</strong></td><td>830</td><td>OpenAI's best, with synced audio and voice.</td><td>Premium AI video generation, cinematic quality, up to 1080p resolution.</td></tr><tr><td><strong>Tencent Hunyuan</strong></td><td>240</td><td>Open-source model, customizable for fine-tuning.</td><td>Multi-locale campaigns, culturally nuanced content, synced soundtracks.</td></tr><tr><td><strong>Topaz Astra 2</strong></td><td>720</td><td>Creative video upscaling to 4K with Astra 2, inventing new detail and texture.</td><td>Creative video upscaling to 4K with Astra 2, inventing new detail and texture.</td></tr><tr><td><strong>Topaz Hyperion 2.5</strong></td><td>180</td><td>Convert SDR and AI-generated video to true HDR with Hyperion 2.5.</td><td>Convert SDR and AI-generated video to true HDR with Hyperion 2.5.</td></tr><tr><td><strong>Topaz Starlight</strong></td><td>240</td><td>Diffusion-based video upscaling with Starlight, rebuilding detail the source has lost.</td><td>Diffusion-based video upscaling with Starlight, rebuilding detail the source has lost.</td></tr><tr><td><strong>VEED Subtitles</strong></td><td>30</td><td>Render subtitles onto videos with selectable style preset and source language.</td><td>Render subtitles onto videos with selectable style preset and source language.</td></tr><tr><td><strong>Veo 2</strong></td><td>180</td><td>Good balance of quality and speed.</td><td>Hero shots, trailers, architecture fly-throughs, luxury product reveals.</td></tr><tr><td><strong>Veo 3</strong></td><td>250</td><td>Native audio with cinematic quality.</td><td>4K trailer sequences, high-end commercials, or narrative teasers.</td></tr><tr><td><strong>Veo 3.1</strong></td><td>305</td><td>Google's best: audio, 1080p, up to 8s clips.</td><td>Premium productions, cinematic sequences, highest-quality outputs for professional use.</td></tr><tr><td><strong>Veo 3.1 (Fast)</strong></td><td>360</td><td>Veo 3.1 quality at lower cost.</td><td>Quick iterations on premium content, cost-effective previews before final renders.</td></tr><tr><td><strong>Veo 3.1 Frames</strong></td><td>250</td><td>Animate between start and end frames.</td><td>Precise keyframe animation, controlled transitions.</td></tr><tr><td><strong>Veo 3.1 Frames (Fast)</strong></td><td>390</td><td>Fast frame-to-frame animation.</td><td>Quick keyframe previews, rapid iteration on transitions.</td></tr><tr><td><strong>Veo 3.1 Frames Lite</strong></td><td>135</td><td>More cost effective frame-to-frame animation.</td><td>Budget-friendly keyframe animation.</td></tr><tr><td><strong>Veo 3.1 Ingredients</strong></td><td>250</td><td>Combine reference images into video.</td><td>Multi-reference video generation, consistent character and style.</td></tr><tr><td><strong>Veo 3.1 Lite</strong></td><td>120</td><td>Veo 3.1 with more cost effective generations.</td><td>Budget-friendly premium video, scaled production workflows.</td></tr><tr><td><strong>WAN 2.2</strong></td><td>120</td><td>Diverse motion styles and camera options.</td><td>Cost-effective realistic image generation</td></tr><tr><td><strong>WAN 2.5</strong></td><td>250</td><td>Built-in audio generation.</td><td>High-quality generation, improved motion, versatile styles.</td></tr><tr><td><strong>WAN 2.6</strong></td><td>250</td><td>Uses audio input to drive the video generation.</td><td>Complex sequences, multi-shot videos, high-resolution outputs.</td></tr><tr><td><strong>WAN 2.7</strong></td><td>250</td><td>Uses audio input to drive the video generation.</td><td>Cinematic multi-shot storytelling with built-in audio generation.</td></tr><tr><td><strong>WAN 3.0</strong></td><td>180</td><td>Up to 30 seconds in one pass with native audio, keyframes, and image, video, and audio references.</td><td>Up to 30 seconds in one pass with native audio, keyframes, and image, video, and audio references.</td></tr></tbody></table>

## Editing & Reference Models

These models edit existing videos or use reference inputs for guided generation.

<table><thead><tr><th width="220">Model</th><th width="120">Est. time (s)</th><th>Description</th><th>Best for</th></tr></thead><tbody><tr><td><strong>Animatediff</strong></td><td>300</td><td>Reanimate video with new motion.</td><td>Rapid motion-graphics, looping GIFs, meme-style clips.</td></tr><tr><td><strong>Fabric 1.0</strong></td><td>60</td><td>Flagship lipsync model using image and audio inputs.</td><td>Flagship lipsync model using image and audio inputs.</td></tr><tr><td><strong>FLUX 3 Video Edit</strong></td><td>240</td><td>Edit a clip from an instruction: add, remove or replace objects, change the setting, restyle, or rewrite dialogue while motion and framing stay put.</td><td>Edit a clip from an instruction: add, remove or replace objects, change the setting, restyle, or rewrite dialogue while motion and framing stay put.</td></tr><tr><td><strong>Grok Imagine Edit</strong></td><td>120</td><td>Edit a video using text prompts.</td><td>Quick edits powered by xAI</td></tr><tr><td><strong>Grok Imagine References</strong></td><td>85</td><td>Generate a video guided by one or more reference images.</td><td>Reference-guided video generation with xAI.</td></tr><tr><td><strong>Happy Horse 1.0 Edit</strong></td><td>250</td><td>Generate videos with synced audio using customizable parameters.</td><td>Generate videos with synced audio using customizable parameters.</td></tr><tr><td><strong>Happy Horse 1.0 References</strong></td><td>290</td><td>Use up to 9 images as references to generate a video with integrated audio.</td><td>Use up to 9 images as references to generate a video with integrated audio.</td></tr><tr><td><strong>Kling 2.6 Pro Motion Control</strong></td><td>96</td><td>Animate character from reference.</td><td>Character animation with 2.6 Pro quality.</td></tr><tr><td><strong>Kling 3.0 Pro Motion Control</strong></td><td>655</td><td>Animate character from reference.</td><td>Character animation with precise motion control.</td></tr><tr><td><strong>Kling 3.0 Standard Motion Control</strong></td><td>560</td><td>Animate character from reference.</td><td>Cost-effective character animation.</td></tr><tr><td><strong>Kling O1 Edit</strong></td><td>450</td><td>Edit video with text or image guidance.</td><td>Precise video editing with multimodal guidance.</td></tr><tr><td><strong>Kling O1 Reference</strong></td><td>450</td><td>Style transfer from a reference video.</td><td>Video style transfer, consistent aesthetic application.</td></tr><tr><td><strong>Kling O1 References</strong></td><td>255</td><td>Keep subjects consistent across scenes.</td><td>Multi-reference consistency, character continuity.</td></tr><tr><td><strong>Kling O3 Pro Edit</strong></td><td>450</td><td>Flagship video editing from Kling.</td><td>Premium video editing with maximum quality.</td></tr><tr><td><strong>Kling O3 Pro Reference</strong></td><td>450</td><td>Use an input video and images as reference.</td><td>Reference-guided generation with highest quality.</td></tr><tr><td><strong>Kling O3 Pro References</strong></td><td>385</td><td>Keep subjects consistent across scenes.</td><td>Multi-reference consistency at premium quality.</td></tr><tr><td><strong>Kling O3 Standard Edit</strong></td><td>450</td><td>Kling O3 at a lower cost and faster speed.</td><td>Cost-effective video editing with O3 capabilities.</td></tr><tr><td><strong>Kling O3 Standard Reference</strong></td><td>450</td><td>Kling O3 at a lower cost and faster speed.</td><td>Budget-friendly reference-guided generation.</td></tr><tr><td><strong>Kling O3 Standard References</strong></td><td>265</td><td>Kling O3 at a lower cost and faster speed.</td><td>Cost-effective multi-reference consistency.</td></tr><tr><td><strong>Lipsync 2 Pro</strong></td><td>300</td><td>Lipsync an input video performance with input speech.</td><td>Lip synchronization, dubbed content, talking-head alignment.</td></tr><tr><td><strong>Lucy Edit Pro</strong></td><td>120</td><td>Text-prompted video editing.</td><td>Quick text-driven video modifications and adjustments.</td></tr><tr><td><strong>Luma Modify Video</strong></td><td>205</td><td>Restyle with adhere, flex, or reimagine modes.</td><td>Day-for-night, illustrative repainting, live-action to anime transformations.</td></tr><tr><td><strong>Marey Motion Transfer</strong></td><td>400</td><td>Apply motion from one video to another.</td><td>Choreography transfer, steadicam paths to new scenes.</td></tr><tr><td><strong>Marey Pose Transfer</strong></td><td>300</td><td>Copy body poses to new subjects.</td><td>Digital-double pose matching, stunt-viz, freeze-frame-to-motion.</td></tr><tr><td><strong>Seedance 2.0 Reference</strong></td><td>390</td><td>Generate video guided by reference images</td><td>Reference-guided video generation with audio.</td></tr><tr><td><strong>Seedance 2.0 Reference (Fast)</strong></td><td>235</td><td>Generate video guided by a single reference image, fast model via Enhancor.</td><td>Quick reference-guided video iteration.</td></tr><tr><td><strong>Sync 3</strong></td><td>300</td><td>Next-gen lipsync model with studio-grade capabilities.</td><td>Next-gen lipsync model with studio-grade capabilities.</td></tr><tr><td><strong>VEED Lipsync</strong></td><td>60</td><td>Generate realistic lipsync using an input audio file and a video.</td><td>Generate realistic lipsync using an input audio file and a video.</td></tr><tr><td><strong>WAN 2.2 Animate Move</strong></td><td>1535</td><td>Apply motion to reference character.</td><td>Character animation from reference, motion transfer.</td></tr><tr><td><strong>WAN 2.2 Animate Replace</strong></td><td>1440</td><td>Swap subjects using reference image.</td><td>Subject replacement in video, face and character swaps.</td></tr></tbody></table>

## Upscaling & Post-Processing

These models enhance, upscale, or post-process existing videos.

<table><thead><tr><th width="220">Model</th><th width="120">Est. time (s)</th><th>Description</th><th>Best for</th></tr></thead><tbody><tr><td><strong>Bria Video Upscaler</strong></td><td>15</td><td>Enterprise-safe, commercially licensed.</td><td>Commercial video upscaling with licensing compliance.</td></tr><tr><td><strong>Magnific Creative Turbo Video Upscaler</strong></td><td>300</td><td>Upscale videos faster while maintaining visual quality.</td><td>Fast video upscaling with creative enhancement.</td></tr><tr><td><strong>Magnific Creative Video Upscaler</strong></td><td>490</td><td>Upscale videos up to 4k resolution.</td><td>Artistic video upscaling to 4K.</td></tr><tr><td><strong>Magnific Precision Video Upscaler</strong></td><td>490</td><td>High-accuracy video upscaling.</td><td>Detail-preserving video upscaling.</td></tr><tr><td><strong>Mirelo SFX 1.5</strong></td><td>30</td><td>Add synced sounds to a video input.</td><td>Add synced sounds to a video input.</td></tr><tr><td><strong>MMaudio v2</strong></td><td>30</td><td>Add sound to your video.</td><td>Automatic audio generation from video content.</td></tr><tr><td><strong>Topaz Upscaler</strong></td><td>165</td><td>Professional video upscaling to 4K.</td><td>Industry-standard upscaling for professional workflows</td></tr><tr><td><strong>VEED Background Removal</strong></td><td>60</td><td>Remove the background from your video.</td><td>Green-screen effects, subject isolation, compositing.</td></tr></tbody></table>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.flora.ai/models/video-models.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.