---
title: Generate video
display: '*Generate* video'
nav_title: Generate video
summary:
  - "Open Create > Videos from the top menu, and choose the model in the model library: search \"last frame\" for models that take a first and a last frame."
  - "Write one shot: what happens, what the camera does, and the sound. First Frame is where the clip starts, Last Frame where it must end."
  - "Set Duration and Resolution yourself: a lower resolution for test runs, then check the price on Generate."
description: "Use the video panel: choose a model in the library, then the prompt, First Frame and Last Frame, references, duration, resolution, aspect ratio and the price before you click."
section: video
order: 10
type: guide
level: Start
roles: [all]
time: { read: 9, do: 10, wait: 5 }
needs: ["Any plan with credits for video; the price shows on the Generate button", "An approved first frame, and an approved last frame for a first and last frame clip"]
last_verified: 2026-11-11
models_checked: ["Minimax H3", "FLUX.3 First Last Frame to Video", "Veo 3.1", "Minimax H3 Fast", "Seedance 2.5", "Seedance 2.5 Draft Complete", "Gemini Omni 1.1 Flash", "Wan 3.0"]
hero: video/generate-video/temple-first-frame.webp
video: tutorials/generate-video.mp4
video_caption: Generate video
about: "This lesson makes a 6-second clip of a temple door opening, from a first frame and a last frame, with Minimax H3 in Create › Videos. You'll pick the model in the model library, write a one-shot prompt, load the two frames, set the duration and resolution, and check the clip. The models and prices in this lesson are as of November 2026."
articles: [generate-video]
chapters:
  - { at: 0, title: "Open Videos and choose a model", image: "/media/images/video/generate-video/ch-01-open-videos-and-choose-a-model.webp" }
  - { at: 37.43, title: "Inputs", image: "/media/images/video/generate-video/ch-02-inputs.webp" }
  - { at: 58.94, title: "Settings", image: "/media/images/video/generate-video/ch-03-settings.webp" }
related: [your-first-video, prompt-video, direct-the-camera-and-motion, reference-to-video, video-models-guide]
thumb: video/generate-video/thumb.webp
steps:
  - title: "Open Videos"
    text: "From the top menu, choose Create › Videos. The panel opens on the left with Model, Prompt, First Frame, Last Frame, the references and Settings; takes appear on the right under Session."
  - title: "Choose a model in the library"
    text: "Click the Model card › View all models, type \"last frame\" in the search to see only the models that take a first and a last frame, pick Minimax H3 and click Use this model."
  - title: "Write the prompt"
    text: "One shot: the look, the action in order, a static camera said plainly, a continuity line, real-time pace and the sound, with what you don't want."
    prompt: |-
      Stylized faceted low-poly 3D game cinematic. Her open hand touches the carved stone sun. A thin seam of warm light appears down the centre of the round door, and the two stone halves slowly slide apart into the wall, in full view, revealing a staircase rising into a golden hall of pillars. Soft warm light spills onto the steps, with drifting dust motes, no bright flash. She lowers her arm and walks up the last steps with small, calm, natural steps, her arms swinging gently, then stops on the top stair at the threshold and stands still with both arms down at her sides, looking into the hall. Static camera: the camera does not move. Same faceted hand-painted style, same character and temple from start to end. Real-time pace. Sound: deep stone rumble and grinding as the door slides, a soft shimmer as the light spills out, jungle birds in the background. No music, no speech.
  - title: "Load the first and last frame"
    text: "Drag the approved first frame onto First Frame and the last frame onto Last Frame. Aspect Ratio follows the first frame; don't mix frames with Reference Images, Videos or Audio in the same run."
  - title: "Know the reference inputs"
    text: "Reference Images (subjects and style, up to 9), Reference Videos (motion, up to 3 clips) and Reference Audio are the other way in. This lesson doesn't use them, since it uses frames."
  - title: "Set duration, resolution and aspect ratio"
    text: "Settings: Duration 6, Resolution 768P for test runs (up to 2K for the final take), Aspect Ratio on Auto. Check the price on the Generate button, then click it."
  - title: "Check the clip"
    text: "Open the take in the viewer and watch it at full size with sound, more than once. Accept when the first and last moments match the frames, the character and place stay the same and the camera stays still; reject cross-fades, a changed face or unwanted music."
agent_steps:
  - title: "Connect the Scenario MCP"
    text: "Connect your agent and sign in once at [mcp.scenario.com](https://mcp.scenario.com). Then tell it which Scenario team and project to use, so every file lands there."
  - title: "Give the agent the files"
    text: "The approved first frame (her hand on the closed door) and the approved last frame (the door open, her on the top stair). Upload them with upload_asset or give their asset IDs, and say which is which."
  - title: "Brief the agent"
    prompt: |-
      Make a 6-second clip with Minimax H3 (model_minimax-h3): firstFrameImage set to the first frame, the last frame in the model's last-frame input (read model_schema_get for its parameter name), duration 6, resolution "768P" for this test take, aspect ratio left to follow the first frame. No reference images, videos or audio in the same run. Quote the cost with a dry run first and wait for my go.

      Prompt, word for word:
      Stylized faceted low-poly 3D game cinematic. Her open hand touches the carved stone sun. A thin seam of warm light appears down the centre of the round door, and the two stone halves slowly slide apart into the wall, in full view, revealing a staircase rising into a golden hall of pillars. Soft warm light spills onto the steps, with drifting dust motes, no bright flash. She lowers her arm and walks up the last steps with small, calm, natural steps, her arms swinging gently, then stops on the top stair at the threshold and stands still with both arms down at her sides, looking into the hall. Static camera: the camera does not move. Same faceted hand-painted style, same character and temple from start to end. Real-time pace. Sound: deep stone rumble and grinding as the door slides, a soft shimmer as the light spills out, jungle birds in the background. No music, no speech.
  - title: "Check and redo"
    text: "Accept when the first and last moments match the two frames, the character and place stay the same, the motion follows the prompt's order, the camera stays still and the sound fits. Reject a jump or cross-fade to reach the last frame, a changed face or clothes, odd hands or feet, a moving camera, or music or speech."
    prompt: |-
      Wait for the job with jobs_wait and show me the clip. Check it against the two frames and the prompt and tell me what fails. If it passes, quote the same run at 2K and wait for my go before the final take.
  - title: "File and report"
    prompt: |-
      Create a collection for this clip with the two frames and the takes, and report the model, the settings and every asset ID.
updated: 2026-11-11
canonical: https://learn.scenario.com/video/generate-video
---

Last updated: 2026-11-11

# Generate video

> Use the video panel: choose a model in the library, then the prompt, First Frame and Last Frame, references, duration, resolution, aspect ratio and the price before you click.

In short:

- Open Create > Videos from the top menu, and choose the model in the model library: search "last frame" for models that take a first and a last frame.
- Write one shot: what happens, what the camera does, and the sound. First Frame is where the clip starts, Last Frame where it must end.
- Set Duration and Resolution yourself: a lower resolution for test runs, then check the price on Generate.


This lesson makes a 6-second clip of a temple door opening, from a first frame and a last frame, with Minimax H3 in **Create › Videos**. You'll pick the model in the model library, write a one-shot prompt, load the two frames, set the duration and resolution, and check the clip.

:::note As of November 2026
The models, settings and prices in this article are the ones Scenario offered in November 2026. New models and updates arrive often, so check the model library for newer options before you choose.
:::

:::steps Quick version
1. From the top menu, choose **Create**, then **Videos**.
2. Click the **Model** card, then **View all models**. Search "last frame", pick **Minimax H3** and click **Use this model**.
3. Write the prompt: what happens, what the camera does, and the sound.
4. Load the **First Frame** and the **Last Frame**.
5. Under **Settings**, set **Duration** and **Resolution**, and leave **Aspect Ratio** on **Auto**.
6. Check the price on the **Generate** button, click it, and watch the clip from start to end.
:::

## Before you start

- **Approve the frames first.** In a first and last frame clip, the model only invents the motion between two pictures you give it. If a frame is wrong, the clip is wrong. Make and approve both frames as images before you open Videos. The last frame is usually an edit of the first, so the character, style and place match.

:::gallery The two frames this lesson uses
![The first frame: a low-poly explorer with her hand on a carved stone sun door at the top of jungle steps](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/a0dfe2dff1e7acd798817c978c5ee2d269ae91ce12d7fa84535b6aaaba21cdab.webp "First frame: her hand on the closed door.")
![The last frame: the same door open on a golden hall, the explorer standing on the top stair](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/88a4b36467c2f8d6a15282261bb14797b2f761a3fdf8f663eab195261123e5f6.webp "Last frame: the door open, her on the top stair.")
:::

- **Know the three ways to drive a video model.** Most models take one or more of these:
  - **Text only:** the prompt alone. Good for open exploration, hard to control.
  - **First and last frame:** the clip starts on one image and, if the model supports it, ends on another. The most control over what the clip shows. This is the lesson's example.
  - **References:** images, videos or audio that tell the model who and what to show, without fixing the first picture. See [Reference to video](https://learn.scenario.com/video/reference-to-video.md).
- **Not every model takes a last frame.** That's why you choose the model by what it accepts, not only by name.
- **Video costs more than images.** The price depends on the model, the duration and the resolution, and it shows on the **Generate** button before you click. Test at a lower resolution, then make the final take.

## Step 1: Open Videos

From the top menu, choose **Create**, then **Videos**. It sits in the **Generate** column, next to **Images**, **3D**, **Audio** and the other generators.

![The Create menu open, with Videos in the Generate column](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/5a8b7c444b9288fff78649305a26beeeb3596d4c66673f4b7b8e04731fc342a7.webp)

The video panel opens on the left: **Model**, **Prompt**, **First Frame**, **Last Frame**, **Reference Images**, **Reference Videos**, **Reference Audio** and **Settings**, with the **Generate** button at the bottom. Your takes appear on the right, under **Session**.

## Step 2: Choose a model in the library

The panel opens with the last model you used. To choose another, click the **Model** card. A short list opens with a search field (**Search by keyword**), filters by output type, and the top-ranked video models: here **Minimax H3**, **Gemini Omni 1.1 Flash**, **Wan 3.0**, **Seedance 2.5** and others. Each line shows a rank, a typical time and a cost level from $ to $$$.

To see every model, click **View all models** at the bottom of the list.

![The Model card open on the video models, with View all models at the bottom](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/4d63ac148360541211fdfdda7d23369bec7f463ce5505a6af89b217be36115a4.webp)

The library opens as **Choose a model**. The filters on the left narrow it by **Output**, **Input** and **Provider**; the **Video** output filter is already on.

To find the models that take a first and a last frame, type **last frame** in the search field. The list shows only those (20 when the film was made), for example **FLUX.3 First Last Frame to Video**, **Minimax H3**, **Veo 3.1**, **Veo 3.1 (Fast)**, **Vidu Q3 Pro I2V**, **Kling V3 I2V Pro**, **Luma Ray 3.2** and **Kling O1 I2V**.

Click a model to read about it in the panel on the right: what it does, its **Recommended** rank, **Typical time**, **Cost** level, and its **Input** and **Output**. For **Minimax H3**, the panel says it generates video from text, first and last frames, or multimodal references, with native stereo audio, and makes 5 to 15 second clips at up to 2K and 24 fps.

![The model library searched for "last frame", with Minimax H3 selected and its description on the right](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/63d5a20e72694a0c9eb1a84c176fe5145a314ab6802e89dcf8a7569ae23f7ef1.webp)

Click **Use this model**. The library closes and **Minimax H3** is now on the **Model** card.

## Step 3: Write the prompt

The prompt describes **one shot**: what happens, what the camera does, and the sound. With a first and a last frame, it describes the motion that carries the first picture to the last one. It doesn't need to describe the place or the character again in detail: the frames already show them.

![The Prompt field with the start of the temple prompt](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/6dd711b0070c66fe9dfd5016714b0634a2e5fa49bf5683f73dfa830952049948.webp)

The prompt the film uses:

```text
Stylized faceted low-poly 3D game cinematic. Her open hand touches the carved stone sun. A thin seam of warm light appears down the centre of the round door, and the two stone halves slowly slide apart into the wall, in full view, revealing a staircase rising into a golden hall of pillars. Soft warm light spills onto the steps, with drifting dust motes, no bright flash. She lowers her arm and walks up the last steps with small, calm, natural steps, her arms swinging gently, then stops on the top stair at the threshold and stands still with both arms down at her sides, looking into the hall. Static camera: the camera does not move. Same faceted hand-painted style, same character and temple from start to end. Real-time pace. Sound: deep stone rumble and grinding as the door slides, a soft shimmer as the light spills out, jungle birds in the background. No music, no speech.
```

What each part does:

- **The look first.** "Stylized faceted low-poly 3D game cinematic" tells the model to keep the style of the frames.
- **The action, in order.** The hand touches the door, light appears, the door opens, she walks up, she stops. Each beat leads to the next, and the last beat matches the last frame.
- **The camera.** "Static camera: the camera does not move." Say it plainly, or many models add a slow push in. To move the camera on purpose, see [Direct the camera and motion](https://learn.scenario.com/video/direct-the-camera-and-motion.md).
- **Continuity.** "Same faceted hand-painted style, same character and temple from start to end" keeps the model from drifting.
- **The pace.** "Real-time pace" avoids slow motion.
- **The sound.** Minimax H3 makes audio in the same pass. Describe it, and say what you don't want: "No music, no speech."

For more on writing video prompts, see [Prompt video](https://learn.scenario.com/methods/prompt-video.md).

## Step 4: Load the first and last frame

**First Frame** is where the clip starts. **Last Frame** is where it has to end: here, the door open and her on the top stair. Drag each image onto its box, or choose it from your library. The buttons on each box let you remove the image, edit it, or open more options.

![First Frame and Last Frame loaded in the video panel](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/09cd065cf9ea43cff568cc6f0c6699c9fdfd6a79f6c12dab3a7d1364be30bc97.webp)

A few rules from the Knowledge Center for Minimax H3:

- The last frame is optional. With only a first frame, the model decides where the clip ends.
- **Aspect Ratio** follows the first frame.
- Don't mix frames with **Reference Images**, **Reference Videos** or **Reference Audio** in the same run. Choose one way in.

## Step 5: Know the reference inputs

**Reference Images**, **Reference Videos** and **Reference Audio** are another way to start. Each has **Drag & drop**, **Select from Library** and **Upload**; audio also has **Record**. This lesson doesn't use them, since it uses frames.

![The Reference Videos, Reference Audio and Settings sections of the video panel](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/78a9fb6ce453c5d9e004201fc61a9948ba251d1d7e8fc490c63014496e3d23ed.webp)

For Minimax H3, the Knowledge Center gives these limits:

| Input | What it guides | Limit |
|---|---|---|
| Reference Images | Subjects and style | Up to 9 images |
| Reference Videos | Motion | Up to 3 clips, 2 to 15 seconds in total |
| Reference Audio | Sound and voice | Only with at least one reference image or video |

Other models take different inputs and limits. The model's panel in the library lists them under **Input**.

## Step 6: Set duration, resolution and aspect ratio

Open **Settings**:

- **Duration** sets how long the clip is, in seconds. The film sets **6**. Minimax H3 makes 5 to 15 seconds.
- **Resolution** sets the size. Minimax H3 goes up to **2K**. For test runs, set a lower one, here **768P**: it's faster and costs less. Make the final take at the resolution you need.
- **Aspect Ratio** stays on **Auto**: the clip takes the shape of the first frame. Change it only for a text-only clip, where there's no frame to follow.

![Settings with Duration at 6, Resolution at 768P and Aspect Ratio on Auto](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/95e8cf682f453bd1f07b5d01da4bc012de76294eb79f30200991f1492bc4aff4.webp)

The **Generate** button shows the price, and it changes as you change the duration and the resolution. Click **Generate**.

## Step 7: Check the clip

The take appears in the session on the right. Click it to open it in the viewer. The details on the right list the model, the prompt and the frames it used, so you can see exactly what made it.

![The finished clip in the viewer, with the model, prompt and frames listed on the right](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/6757a9af57432c7872afa7a7b28d7d903abee194f4bc7ad4e2842f8febdb1579.webp)

The clip starts on the first frame and ends on the last:

:::video https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/29248a83ddb1903bf1900e6597d523da5596b4d239121a2b62e8302d19d9cee4.mp4
Minimax H3, 6 seconds at 2K. The door opens, she climbs the last steps and stops on the top stair, as in the last frame.
:::

Watch it at full size, with the sound on, more than once: once for the motion, once for the faces and hands, once for the sound.

:::check
**Accept when:** the first and last moments match the two frames, the character and the place stay the same from start to end, the motion follows the order in the prompt, the camera does what you asked, and the sound fits the action.
**Reject when:** the clip jumps or cross-fades to reach the last frame, the character changes face or clothes, hands or feet bend in odd ways, the camera moves when you asked for a static shot, or there's music or speech you said you didn't want.
:::

## Choose a model for first and last frame

Every model in the "last frame" search can make this kind of clip. They differ in length, resolution, sound and speed, and the panel on the right of the library sums each one up. The Minimax family, from the Knowledge Center:

| Model | Inputs | Length and size | Good for |
| --- | --- | --- | --- |
| Minimax H3 | Text, first and last frame, or up to 9 references | 5 to 15 s, up to 2K, 24 fps, native audio | Story beats with sound or dialogue, brand films, game trailers |
| Minimax H3 Max I2V | A starting image, optional end image | 5 to 15 s, 480P or 768P, with audio | Faster tests of a still, with control over how the shot ends |
| Minimax H3 Max Turbo I2V | A starting image, optional end image | 480P or 768P | The fastest way to animate a still when speed matters more than 2K |

To compare models on the same frames, see [Video models](https://learn.scenario.com/video/video-models-guide.md).

## Common problems

| What you see | What to do |
|---|---|
| The model you want isn't in the short list | Click **View all models** and search for it by name or by what it takes, such as "last frame" |
| There's no **Last Frame** box | The model doesn't take one. Search "last frame" in the library and pick one that does |
| The clip cross-fades or cuts to reach the last frame | The two frames are too far apart. Make the last frame closer to the first (same angle, same place), or make the clip longer |
| The camera moves although you didn't ask | Write "Static camera: the camera does not move." in the prompt |
| The character or style changes during the clip | Add a continuity line, such as "same character and style from start to end", and check that both frames match |
| The clip is in slow motion | Add "Real-time pace" to the prompt |
| Unwanted music or voices | Say what you don't want: "No music, no speech." |
| The result has the wrong shape | Leave **Aspect Ratio** on **Auto** so it follows the first frame, and check the first frame's shape |
| The run fails with frames and references | Use frames or references, not both, in the same run |

## What's next

To drive the clip with pictures of a character and a place instead of fixed frames, go to [Reference to video](https://learn.scenario.com/video/reference-to-video.md). To move the camera on purpose, see [Direct the camera and motion](https://learn.scenario.com/video/direct-the-camera-and-motion.md), and to choose between models, [Video models](https://learn.scenario.com/video/video-models-guide.md).

Source: https://learn.scenario.com/video/generate-video