Vidu Q4 AI Video Generator
Turn your own photos into videos in minutes · Your first video is free, no watermark
First video free, no card
Vidu Q4 is the newest video model from Vidu, released on Vidu's API as Vidu Q4 Preview on October 7, 2026.
On RefVid it works as a reference-to-video generator: add 1 to 10 photos of a product, a person or a pet, write what should happen, and Vidu Q4 makes a 3- to 16-second clip with the subject from your photos. The box above is already set to Vidu Q4, and your first clip is free, with no watermark.
Vidu Q4 on RefVid: specs
| Spec | Vidu Q4 on RefVid |
|---|---|
| Clip length | 3–16 seconds |
| Resolution | 720p or 1080p |
| Shapes | 16:9, 9:16, or 1:1 |
| Sound | On or off, same price |
| Reference photos | 1–10 photos of one subject |
| Video from a prompt alone | No, it needs photos |
| End frame (a picture the clip ends on) | No |
| Photos of real-looking people | Yes |
| Longest prompt | 4,700 characters |
| Credits per second, 720p | 25, with or without sound |
| Credits per second, 1080p | 31, with or without sound |
What Vidu Q4 Preview is
- Vidu Q4 Preview is the model Vidu listed on its API on October 7, 2026. Its reference-to-video mode takes up to 15 reference images and up to 3 reference voice clips of 3 to 12 seconds each, and makes clips of 3 to 16 seconds at 540p, 720p, 1080p, 2K or 4K, according to Vidu's API documentation.
- RefVid runs Vidu Q4 through fal.ai, whose endpoint takes up to 12 reference images and a prompt of up to 5,000 characters. In the RefVid box you add up to 10 photos of one subject, pick 720p or 1080p and switch sound on or off; sound costs nothing extra on Vidu Q4. Vidu Q4 needs photos: for a clip from words alone, pick another model in Settings.
Vidu Q4 vs Vidu Q3
- Q4 adds 2K and 4K output and up to 15 reference images, where Q3 took 7, according to Vidu's API docs. It also takes up to 3 reference voice clips, which base Q3 does not; Vidu's Q3 Mix and Q3 Ad variants take 1 to 3. The clip length stays at up to 16 seconds. On fal, Q4 also costs less per second than Q3: $0.0665 at 720p and $0.084 at 1080p, against $0.154 for Q3 at both.
| Vidu Q3 | Vidu Q4 Preview | |
|---|---|---|
| Released | January 27, 2026 (Q3 Pro); reference to video on April 13, 2026 | October 7, 2026 |
| Clip length | 1–16 seconds | 3–16 seconds |
| Resolutions | 540p, 720p, 1080p | 540p, 720p, 1080p, 2K, 4K |
| Reference images (Vidu API) | 1–7 | 1–15 |
| Reference images (fal) | 1–4 | Up to 12 |
| Reference voice clips | None on base Q3; 1–3 on Q3 Mix and Ad | Up to 3, 3–12 seconds each |
| Price per second on fal (720p / 1080p) | $0.154 / $0.154 | $0.0665 / $0.084 |
Vidu Q4 reference to video: one subject, every clip
- Reference to video means the model reads your photos as a description of who or what must appear, then builds a new scene around it. With Vidu Q4 on RefVid, the photos you add are saved as one subject, a product or a character, so the next clip starts from the same photos without uploading them again.
- Three to five photos from different angles give the model the most to work with: front, three-quarter and back for a product, plus a close-up of the label; front, profile and a full-body shot for a person. Plain light and a clean background help more than a busy scene.
- Write the scene, not the subject: where it is, what moves, how the camera moves. For example: "The bottle turns slowly on a white marble plinth, morning light sweeps across the label, the camera pushes in to a close-up." The subject's look comes from your photos.
Make a Vidu Q4 video in three steps
- Add 1 to 10 photos of your product or character in the box at the top of this page, or start from a template in the row above it. The box is already set to Vidu Q4, 720p and the free clip's length.
- Write what should happen in one or two sentences, or keep the template's words. Settings changes the length (3 to 16 seconds), the shape (16:9, 9:16 or 1:1), 1080p and sound.
- Press Generate. If you are not signed in, you sign in first and the clip starts right after. A clip usually takes 2 to 10 minutes, and we email you when it is ready.
What a Vidu Q4 clip costs on RefVid
- Every clip is priced in credits per second, as in the table above, and the studio shows the total before anything runs. A new account's free credits cover its first clip. After that, single videos start at $2.99, one-time credit packs at $9.90 and monthly plans at $10. A failed clip refunds its credits automatically.
Vidu Q4 next to every RefVid model: measured clips, wait and cost
- These numbers are our own, not the vendor's: every clip people outside our team started on RefVid from October 27 – November 6, 2026, split by the model they picked, with how many finished and the median wait from the start of the generation to the finished clip. Next to them are each model's photo limit, clip length and what a 5-second clip costs today, so Vidu Q4 can be read against the other models in one table.
- The finished share and the wait are shown only where a model had at least 20 finished or failed clips in the period; the models added most recently have fewer so far. A failed clip costs nothing: its credits go back to the account. The table follows the newest snapshot of our usage statistics.
| Model | Clips (finished or failed) | Finished | Median wait | Photos of one subject | Clip length | Credits for a 5-second clip | Cost with the $9.90 pack |
|---|---|---|---|---|---|---|---|
| Kling 4 | 18 | Under 20 clips | Under 20 clips | 10 | 3–30 sec | 290 | $1.91 |
| Kling 3.0 Omni | 310 | 83% | 2.1 min | 10 | 3–15 sec | 155 | $1.02 |
| Seedance 2.5 | 3 | Under 20 clips | Under 20 clips | 10 | 4–30 sec | 860 | $5.68 |
| Wan 3.0 | 13 | Under 20 clips | Under 20 clips | 10 | 3–30 sec | 185 | $1.22 |
| Vidu Q4 | 21 | 71% | 4.2 min | 10 | 3–16 sec | 125 | $0.83 |
| Gemini Omni Flash | 1 | Under 20 clips | Under 20 clips | 10 | 3–10 sec | 185 | $1.22 |
| PixVerse C1 | 80 | 99% | 1.4 min | 7 | 3–15 sec | 95 | $0.63 |
| Happy Horse 1.1 | 1 | Under 20 clips | Under 20 clips | 9 | 3–15 sec | 255 | $1.68 |
| MiniMax H3 | 1 | Under 20 clips | Under 20 clips | 5 | 5–15 sec | 110 | $0.73 |
When another model fits better
- Vidu Q4 suits short product and character clips at a low price per second. For clips longer than 16 seconds, Kling 4, Wan 3.0 and Seedance 2.5 reach 30. For a clip from words alone, without photos, pick any other model in Settings. Your saved subjects work with every model, so switching costs nothing.
Start from a template
Each one opens the studio with the shots, shape and length already set. Add your own product and generate.
More on AI video models
Text to video AI →
Frequently asked questions
Is Vidu Q4 free on RefVid?
Your first clip is free: a new account gets 180 credits, enough for a short silent 720p Vidu Q4 clip, with no watermark and no card. After that you pay per second in credits; the studio shows what a clip costs before it runs.
What is the Vidu Q4 release date?
Vidu listed Vidu Q4 Preview on its API on October 7, 2026, according to Vidu's API updates page. fal.ai, which RefVid runs it through, offers it as a reference-to-video endpoint with up to 12 reference images.
How long can a Vidu Q4 video be?
From 3 to 16 seconds in one generation, on Vidu's API and on RefVid. On RefVid you pick the length in Settings, at 720p or 1080p, in 16:9, 9:16 or 1:1.
Does Vidu Q4 do reference to video?
Yes. Vidu's API takes up to 15 reference images and fal's endpoint up to 12. On RefVid you add up to 10 photos of one subject, a product or a character, and it is saved so the next clip starts from the same photos.
Does Vidu Q4 make sound?
Yes, it can generate audio with the clip. On RefVid sound is a switch in Settings and costs the same as a silent clip on Vidu Q4. Vidu's own API also takes up to 3 reference voice clips; RefVid does not pass voice clips yet.