Text-only generation
Four models create video from a prompt alone: Minimax Hailuo 3, Hailuo 3 Max, Luma Ray 3.2 and Grok Imagine Video 1.5.
Describe a scene in plain words and get a video back, with motion, camera work and sound. No footage, no editing software, no filming.
27 video models in one place. Prompt helper built in. Clips with native audio.
Pixara's text to video AI turns a written prompt into a short video. You describe the subject, the action, the setting and the camera, and the model generates the shot, often with synced sound.
Not sure how to phrase it? Write a rough idea and press Enhance My Idea. Pixara rewrites it into a detailed video prompt you can edit before you generate.
Choose Hailuo 3, Hailuo 3 Max, Luma Ray 3.2 or Grok Imagine Video 1.5. On these, the image slot is optional, so a prompt is all you need.
Describe who or what is on screen, what happens, where it happens and how the camera moves. Use Enhance My Idea to add detail.
Pick the aspect ratio (9:16 for Reels and TikTok, 16:9 for YouTube), the length and quality, then generate and download.
Four models create video from a prompt alone: Minimax Hailuo 3, Hailuo 3 Max, Luma Ray 3.2 and Grok Imagine Video 1.5.
Hailuo 3, Hailuo 3 Max and Grok Imagine Video 1.5 generate audio with the video, so the clip is not silent.
16:9, 9:16, 1:1, 4:3, 3:4 and cinematic 21:9, depending on the model. Hailuo 3 and 3 Max also offer Adaptive.
Luma Ray 3.2 offers 5 or 10 second clips at up to 1080p, with a Loop option for seamless background videos.
Enhance My Idea expands a short idea into a full, structured video prompt.
Add a voiceover, sound effects or music, or translate the video into another language without leaving Pixara.
These are the models that accept a prompt with no image (checked in Create Video on 19 Oct 2026).
| Model | Best for | What you can set |
|---|---|---|
| Minimax Hailuo 3 Max | Sharp prompt adherence, realistic people and motion | 480p or 768p, 21:9 to 9:16 or Adaptive, native sound |
| Minimax Hailuo 3 | High-detail video from text, frames or references | 21:9 to 9:16 or Adaptive, native sound |
| Luma Ray 3.2 | Cinematic camera motion and seamless loops | 5 or 10 s, 540p to 1080p, Loop, 21:9 to 9:16 |
| Grok Imagine Video 1.5 | Lively, playful motion with sound | 480p or 720p, eight aspect ratios, native sound |
Want Kling 3.0, Veo 3.1 or Seedance 2.5? These start from a first frame. Generate that frame from your prompt in Create Image, then animate it in Create Video. It takes one extra click and gives you access to all 27 video models.
Formula: Subject + action + setting + camera + light and mood + sound
A golden retriever puppy chases soap bubbles across a sunny backyard lawn, low tracking shot at the dog's eye level, warm afternoon light, playful, the sound of giggling children.
Close-up of a barista pouring latte art in a busy cafe, slow push-in, shallow depth of field, morning light through the window, soft chatter and an espresso machine hissing.
Aerial drone shot rising over a misty pine forest at sunrise, slow forward glide, cinematic, birdsong.
A sneaker spins on a glossy black pedestal, studio lighting with a blue rim light, smooth 360 orbit, bass-heavy beat.
Hooks, B-roll and short scenes for Reels, TikTok and YouTube Shorts without filming.
Concept videos and product spins to test ideas before a real shoot.
Turn a script into scenes for faceless channels, stories and lessons.
Show a client or a team what a shot will look like instead of describing it.
Each clip lasts a few seconds. Describe one clear action per prompt instead of a whole story, and build the story from several clips.
Start with who or what is on screen, then add the action. Models follow the first part of a prompt most closely.
Use simple camera words: close-up, wide shot, slow push-in, tracking shot, orbit, drone shot, handheld.
"Golden hour", "neon-lit street at night" or "soft studio light" changes the whole feel of a clip.
On models with native sound, describe it: crowd noise, rain, a voice laughing, an upbeat beat.
If a clip is close, change one detail in the prompt and generate again rather than writing a new prompt from scratch.
Text to video gives you the shots. Pixara has the rest of the toolkit in the same account, so you can finish the video without exporting to another app.
Write a prompt for each shot of your script and generate them one by one in Create Video.
Bring the clips together in the Pixara video editor and put them in order.
Create a voiceover with ElevenLabs or Minimax voices, add scene-aware sound effects, or generate music for the background.
Use Video Translate to dub the finished video into other languages.
Run your best clips through the AI video upscaler for 1080p, 2K or 4K output.
Start with text to video when you are exploring ideas, need B-roll fast, or do not have an image yet. It is the quickest way from idea to moving picture.
Switch to image to video when the look has to be exact: a specific product, a brand style or the same character across several clips. Generate or upload the image first, then animate it. That route also gives you every model on Pixara, including Kling 3.0, Veo 3.1 and Seedance 2.5.