---
title: Pick a video model by inputs, length, audio and price
display: '*Pick a* video model by inputs, length, audio and price'
nav_title: Video models guide
summary:
  - "Open the model library from the Model card, filter by Input, Budget and Speed or Capabilities, and read each model's rank, typical time and cost."
  - "Compare two or three models on one of your shots, with the same image, the same prompt and the same resolution."
  - "Read the price on Generate at the settings you will ship: duration and resolution change it."
description: How to find a video model in the model library, the video model families as of November 2026, how their prices compare, and the length and resolution each one allows.
section: video
order: 50
type: guide
level: Build
roles: [all]
time: { read: 8, do: 15, wait: 10 }
needs: ["One approved first frame and one shot prompt to compare on"]
last_verified: 2026-11-11
models_checked: ["Seedance 2.5", "Seedance 2.5 Draft Complete", "Kling V4 I2V", "Wan 3.0", "FLUX.3 Max Image to Video", "Minimax H3", "Minimax H3 Fast", "Minimax H3 Max I2V", "Gemini Omni 1.1 Flash", "Gemini Omni 1.2 Flash", "Grok Imagine Video 1.5", "Kling V3 I2V Pro", "Kling V3 Pro - Motion Control", "Vidu Q4 Reference2V", "Kling O1 Reference Images", "Veo 3.1"]
hero: video/video-models-guide/nine-takes-720p-grid.webp
video: tutorials/video-models-guide.mp4
video_caption: Choose a video model
about: "There are hundreds of video models in Scenario, and the right one depends on what you start from, how long you can wait and what you want to spend. This lesson finds models in the model library with its filters, then compares a few of them on the same image and the same prompt, so you can choose on your own shot instead of on a sample. The models and prices in this lesson are as of November 2026."
chapters:
  - { at: 0, title: "Find a model", image: "/media/images/video/video-models-guide/ch-01-find-a-model.webp" }
  - { at: 67.95, title: "Compare models", image: "/media/images/video/video-models-guide/ch-02-compare-models.webp" }
related: [generate-video]
thumb: video/video-models-guide/thumb.webp
steps:
  - title: "Open the model library"
    text: "In Create › Videos, click the Model card › View all models. Choose a model opens already filtered to Video under Output."
  - title: "Filter by what you start from"
    text: "Under Input, tick Text, Image, Video or Audio. Each filter shows as a chip by the count; Clear filters starts again."
  - title: "Read a row, then a model"
    text: "Each row shows the rank (lower is a stronger pick), the typical time and the cost from $ to $$$. Point at a model for its details on the right, then View model page or Use this model."
  - title: "Narrow the list"
    text: "Set Budget ($, $$, $$$) and the Speed slider, pick a task under Capabilities (Lipsync, Video Upscale, Show all 22), or use New · last 30 days under Shortlists; Also show adds Scenario LoRAs and Legacy models."
  - title: "Search and sort"
    text: "Search by name or provider (\"kling\" finds 20 video models), and sort with the button at the top right: Recommended, Recently used, Name (A to Z) or Cost: low to high."
  - title: "Compare models on the same shot"
    text: "Run two or three candidates on the same start image, the same prompt word for word, the same resolution (720p here) and the same duration, then watch every take to the end. The film's three test prompts: a push-in on a face, a time-lapse street, a robot in action."
    prompt: |-
      He slowly turns his head toward the camera and looks straight into the lens, takes a breath, and his eyes narrow slightly into a quiet, determined expression. The wind lifts his hair and the edge of his hood; behind him the red banners ripple and the mist drifts through the ring gate. Slow push-in from a medium close-up to a close-up. Real-time pace. Sound: steady mountain wind, banners flapping. No music, no speech.

      Time-lapse. The woman in the camel coat stands perfectly still with her eyes closed while the crowd streams around her in fast blurred streaks, umbrellas flowing past; the blue-hour sky darkens into night, more shop lights come on, and headlights streak past behind. Locked-off static camera; the woman stays sharp and almost motionless, only her scarf moves slightly. Sound: muffled city ambience, light rain, distant traffic. No music, no speech.

      The little brass robot grips the big iron gear with both hands and pushes it; the gear creaks and slowly begins to turn, the other wheels and the pendulum start moving, and dust shakes loose and drifts through the beams of golden light. The robot steps back and looks up at the turning gears, its amber porthole glowing brighter. Slow camera dolly to the right around the robot. Sound: creaking iron, a heavy clunk as the mechanism starts, ticking. No music, no speech.
  - title: "Choose, and check the price"
    text: "Pick the model that holds the subject and style and moves as asked. Set Duration and Resolution to what you will ship, then read the price on the Generate button."
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 shot to test"
    text: "One approved start image per test (here the hooded man at the ring gate, the woman in the camel coat in the street, the brass robot at the gear), each with its prompt. Upload the images with upload_asset or give their asset IDs."
  - title: "Ask for candidates"
    prompt: |-
      Use models_list to find video models that take an image as input. For each candidate, tell me what it takes, the lengths and resolutions it makes and whether it makes sound. Include Seedance 2.5 (model_bytedance-seedance-2-5), Wan 3.0 (model_alibaba-wan-3-0), Kling V4 I2V, FLUX.3 Max I2V and Minimax H3 Fast (find their ids with models_list). Don't run anything yet.
  - title: "Brief the agent: the fair test"
    prompt: |-
      Run each test on Seedance 2.5 and Kling V4 I2V, with Wan 3.0 on the first two tests and FLUX.3 Max I2V on the robot, plus Minimax H3 Fast at 768P since it has no 720p. Same start image, the prompt word for word, 720p, the same duration where the models allow it (read model_schema_get for each model's parameter names). Quote the cost of every run with a dry run first and wait for my go.

      The face, prompt word for word:
      He slowly turns his head toward the camera and looks straight into the lens, takes a breath, and his eyes narrow slightly into a quiet, determined expression. The wind lifts his hair and the edge of his hood; behind him the red banners ripple and the mist drifts through the ring gate. Slow push-in from a medium close-up to a close-up. Real-time pace. Sound: steady mountain wind, banners flapping. No music, no speech.

      The street, prompt word for word:
      Time-lapse. The woman in the camel coat stands perfectly still with her eyes closed while the crowd streams around her in fast blurred streaks, umbrellas flowing past; the blue-hour sky darkens into night, more shop lights come on, and headlights streak past behind. Locked-off static camera; the woman stays sharp and almost motionless, only her scarf moves slightly. Sound: muffled city ambience, light rain, distant traffic. No music, no speech.

      The robot, prompt word for word:
      The little brass robot grips the big iron gear with both hands and pushes it; the gear creaks and slowly begins to turn, the other wheels and the pendulum start moving, and dust shakes loose and drifts through the beams of golden light. The robot steps back and looks up at the turning gears, its amber porthole glowing brighter. Slow camera dolly to the right around the robot. Sound: creaking iron, a heavy clunk as the mechanism starts, ticking. No music, no speech.
  - title: "Check and choose"
    text: "Accept a model that takes your inputs, reaches your length and resolution, holds the subject and style from start to end and moves as the prompt says. Reject faces or objects that change, a camera that moves when asked to stay still (or the reverse), an action that stops halfway, or a model that can't reach your delivery size."
    prompt: |-
      Wait for the jobs with jobs_wait and show me the takes side by side, test by test. For each take, tell me where the subject or style drifts, whether the camera does what the prompt asked, and whether the action finishes, watching to the last second. Then quote the winner at the duration and resolution I will ship.
  - title: "File and report"
    prompt: |-
      Create a collection for this comparison with the start images and every take, and report the models, the settings and every asset ID.
updated: 2026-11-11
canonical: https://learn.scenario.com/video/video-models-guide
---

Last updated: 2026-11-11

# Pick a video model by inputs, length, audio and price

> How to find a video model in the model library, the video model families as of November 2026, how their prices compare, and the length and resolution each one allows.

In short:

- Open the model library from the Model card, filter by Input, Budget and Speed or Capabilities, and read each model's rank, typical time and cost.
- Compare two or three models on one of your shots, with the same image, the same prompt and the same resolution.
- Read the price on Generate at the settings you will ship: duration and resolution change it.


There are hundreds of video models in Scenario, and the right one depends on what you start from, how long you can wait and what you want to spend. This lesson finds models in the model library with its filters, then compares a few of them on the same image and the same prompt, so you can choose on your own shot instead of on a sample.

:::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. In **Create › Videos**, click the **Model** card, then **View all models**.
2. Under **Input**, tick what you start from: **Text**, **Image**, **Video** or **Audio**.
3. Narrow with **Budget and speed**, **Capabilities** or **New · last 30 days**, or search by name or provider.
4. Run two or three models on the same image and prompt at the same resolution.
5. Pick the one that holds up, and read the price on **Generate** at the duration and resolution you will ship.
:::

## Before you start

- **Have one shot ready to test.** An approved image to start from and one prompt for it. Comparing models on your own shot tells you more than any sample.
- **Know what you start from.** A prompt only, a still to animate, a clip to edit or extend, or a sound track to follow. That is the first filter, and it removes most models.
- **Know where it is going.** The length and resolution you need rule out some models, and both change the price.
- **Where the library lives.** Open **Create › Videos** from the top menu. The **Model** card sits at the top of the panel on the left. The same library opens from every generation panel.

## Step 1: Open the model library

Click the **Model** card. A short list opens: the top video models with their rank, typical time and cost, a search field, and buttons to filter by type (**ALL**, image, video, audio, 3D, world and text). At the bottom, click **View all models**.

![The Model card open, with Kling V4 I2V selected, the short list of video models and View all models at the bottom](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/f2c42ab2217b704b87e22e16a9a665010c9817b93851adc0ef04a55b885b5cc9.webp)

The library opens as **Choose a model**: filters on the left, the list in the middle and the details of the highlighted model on the right. Opened from **Videos**, it is already filtered to **Video** under **Output**.

## Step 2: Filter by what you start from

Under **Input**, tick what you are starting from:

- **Text**, to make a video from a prompt only.
- **Image**, to animate a still, or to use reference images.
- **Video**, to edit, restyle or extend a clip.
- **Audio**, to drive the video with a sound track.

The count above the list updates as you tick, and each filter shows as a chip next to it (for example **Input: Text**). Click **Clear filters** to start again.

![The model library filtered to Video output and Text input, with 40 models in the list](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/4f8acee01140e08077f0524fbe98a3497678878a620e1d72efe8993f83dc0ba9.webp)

## Step 3: Read a row, then a model

Each row shows the model's name, its provider, its **rank**, its **typical time** and its **cost**. The small icons under the name show what goes in and what comes out (for example image to video). Tags mark **New** models and models that are **Model in Preview**.

- **Rank** is the model's place in Scenario's recommendations. A lower number is a stronger pick.
- **Typical time** is how long a generation usually takes, from seconds to several minutes.
- **Cost** is shown as **$**, **$$** or **$$$**, so you can compare models at a glance. The exact price for your settings shows on the **Generate** button.

Point at a model to read its details on the right: **Recommended**, **Typical time** and **Cost**, a short description, its **Input** and **Output**, and **Content samples**. **View model page** opens its full page, and **Use this model** loads it into the panel.

![The details of Gemini Omni 1.1 Flash: rank #4, typical time 42s, its description, and Input Text / Image, Output Video](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/3b8ea2455e4ed89304bd9bbfd974c58da1517690e143c4f3f489bffb755e52a1.webp)

## Step 4: Narrow the list

There are other ways to narrow the list.

**Budget and speed.** Under **Budget**, click **$**, **$$** or **$$$** to set the most you want to spend (**Maximum cost**). Under **Speed**, drag the slider to set how long you can wait, from **10s** to **Any speed**.

![Budget and speed: the Budget buttons and the Speed slider, under Shortlists](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/f04ef340d7d1461932f2ea0023f949b4ed7bd46adc353f542274c2d7403c035d.webp)

**Capabilities** finds models for a task, like **Lipsync** or **Video Upscale**. The list also has **Tools**, **Animate**, **Sound Effects**, **Dubbing**, **Speech**, **Expand**, **Music** and more. Click **Show all 22** to see every one.

![The Capabilities filter, with Tools, Lipsync, Animate, Video Upscale and the other tasks](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/ad2735e3cd94115aa0340b12d269df99260766831004531cdadf500f7560b05a.webp)

**New · last 30 days**, under **Shortlists**, shows the latest models. Under **Also show**, you can add **Scenario LoRAs** and **Legacy models** to the list.

![New · last 30 days ticked, with the newest video models in the list](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/1600fac9f3c5fe2da16b3016bbac7a77bdc77abe355b7a4a5497f874fb96948c.webp)

## Step 5: Search and sort

Type in the search field (**Search 509 models by name or provider** in the film) to find a model or a family. Here, "kling" finds 20 video models, from Kling V4 I2V to the Kling V3 motion-control models.

![A search for "kling" returning 20 video models](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/4dbc9653a78b637afe0e8ebe52f57cf3278e1fa778377ee18d79aa1cb480ceaf.webp)

The button at the top right of the library sorts the list: **Recommended** (the default), **Recently used**, **Name (A to Z)** or **Cost: low to high**.

![The Sort by menu: Recommended, Recently used, Name (A to Z), Cost: low to high](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/e270c85cf99f9dc3e2cbe58aafbd6a93fe357a175a5ca8644c7ebbfffbcefa3c.webp)

The keys at the bottom of the library work too: the arrow keys move through the list, Enter selects, / jumps to search and Esc closes.

## Step 6: Compare models on the same shot

A model that looks best in its samples may not be the best on your shot. Run two or three candidates with the same image, the same prompt and the same resolution, then watch the takes side by side.

The film runs three tests at 720p, each with one start image and one prompt.

![The start image of the first test: a man in a gray hooded cloak in front of a stone ring gate with red banners](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/e997cccd96160e652d3c2567dfa3627f6999d31594a76a36001d27eb77037d45.webp)

A slow push-in on a face:

```text
He slowly turns his head toward the camera and looks straight into the lens, takes a breath, and his eyes narrow slightly into a quiet, determined expression. The wind lifts his hair and the edge of his hood; behind him the red banners ripple and the mist drifts through the ring gate. Slow push-in from a medium close-up to a close-up. Real-time pace. Sound: steady mountain wind, banners flapping. No music, no speech.
```

A time-lapse in a busy street:

```text
Time-lapse. The woman in the camel coat stands perfectly still with her eyes closed while the crowd streams around her in fast blurred streaks, umbrellas flowing past; the blue-hour sky darkens into night, more shop lights come on, and headlights streak past behind. Locked-off static camera; the woman stays sharp and almost motionless, only her scarf moves slightly. Sound: muffled city ambience, light rain, distant traffic. No music, no speech.
```

A character in action:

```text
The little brass robot grips the big iron gear with both hands and pushes it; the gear creaks and slowly begins to turn, the other wheels and the pendulum start moving, and dust shakes loose and drifts through the beams of golden light. The robot steps back and looks up at the turning gears, its amber porthole glowing brighter. Slow camera dolly to the right around the robot. Sound: creaking iron, a heavy clunk as the mechanism starts, ticking. No music, no speech.
```

Each test ran on **Seedance 2.5** and **Kling V4 I2V**, with **Wan 3.0** on the first two and **FLUX.3 Max I2V** on the robot. **Minimax H3 Fast** has no 720p option, so its takes are shown at 768P.

Keep the comparison fair:

- Same start image, same prompt, word for word.
- Same resolution, or the nearest one when a model doesn't offer it.
- Same duration where the models allow it.
- Watch every take to the end. Problems often show in the last second.

## Step 7: Choose, and check the price

Based on what you need, choose the model you want. For example, if you want to test ideas at a lower cost, a fast model like Minimax H3 Fast costs far less per clip than Wan 3.0, and you can move to a stronger model once the idea works.

The price shows on the **Generate** button, and it changes with **Duration** and **Resolution**. Set both to what you will ship before you read it.

![The panel settings for Kling V4 I2V: Start Image, End Image, Duration at 10 and Resolution at 1080p](https://ufgc0c9mhl4ueq76.public.blob.vercel-storage.com/academy/d054e9732fd012d7cbe3efb339b0c5d154632e5d7ea02edd23fa3002e3723399.webp)

:::check
**Accept when:** the model takes the inputs you have, makes the length and resolution you need, holds your subject and style from start to end, and moves the way the prompt says, at a price that fits the number of takes you expect to run.
**Reject when:** faces or objects change during the clip, the camera moves when you asked for a static shot (or the reverse), the action stops halfway, or the model can't reach your delivery size. Try the next model on the same shot before you rewrite the prompt.
:::

## The models in the film

Ranks, times and costs in the library change as models are added, so read them there. This is what each model in the film does, from its Knowledge Center page.

| Model | Starts from | Length and resolution | Notes |
|---|---|---|---|
| Seedance 2.5 | Prompt, first frame, first and last frame, reference images or videos, a clip to edit or extend | 4 to 30 s, up to 1080p | One model for many jobs; it picks the mode from what you attach. Optional native sound |
| Kling V4 I2V | An image (**Start Image**, optional **End Image**) | 1080p in the film's panel | Animates a still; has **Multi Prompt** for timed beats |
| Wan 3.0 | Prompt, first frame with optional last frame, or up to 10 images, 5 videos and 5 audio clips | 2 to 30 s, up to 1080P | Picture and sound in one pass, including spoken dialogue |
| FLUX.3 Max I2V | An image | 5 to 20 s, up to 1080p (FLUX.3 Video family) | Native synchronized audio in the same pass |
| Minimax H3 | Prompt, first and last frame, or up to 9 references | 5 to 15 s, up to 2K | Native stereo audio and dialogue |
| Minimax H3 Fast | See its details in the library | 768P in the film | A low-cost way to test ideas before a stronger model |
| Gemini Omni 1.1 Flash | Prompt, optional first and last frame, up to 7 reference images | 360p to 4K | Native audio. It stops working on December 3, 2026: move to Gemini Omni 1.2 Flash |
| Grok Imagine Video 1.5 | Prompt or one image | Up to 15 s, 480p to 1080p | Native audio, strong faces, one camera move per clip works best |

For the full list of what each model takes and makes, see the [model directory](https://learn.scenario.com/models).

## Choose by the job

- **You only have a prompt.** Filter **Input** to **Text**. Models with native audio (Wan 3.0, Minimax H3, Grok Imagine Video 1.5) can speak lines you write in the prompt.
- **You have an approved still.** Filter to **Image**. This is the most reliable way to keep a look: make the frame first, then animate it.
- **You need the clip to end on a set frame.** Search "last frame" and pick a model that takes a first and a last frame. See [Generate video](https://learn.scenario.com/video/generate-video.md).
- **You need the same character across shots.** Look for reference-to-video models. See [Reference to video](https://learn.scenario.com/video/reference-to-video.md).
- **You need a task, not a generation.** Use **Capabilities**: **Lipsync**, **Video Upscale**, **Dubbing**, **Sound Effects**.
- **You are still exploring.** Use a fast, low-cost model at a lower resolution, then rerun the winner at full quality.

## Common problems

| What you see | What to do |
|---|---|
| The model you want isn't in the list | Check the filters: clear **Input**, **Budget** or **Speed**, or tick **Legacy models** under **Also show** |
| The resolution you need isn't offered | Pick the nearest one to compare, then upscale the chosen take, or filter for a model that reaches it |
| The price on Generate is higher than expected | Lower **Duration** or **Resolution** for test runs; check the price again before the final run |
| Takes from two models are hard to compare | Run them with the same image, prompt, resolution and duration |
| The subject changes during the clip | Try another model on the same shot, simplify the background, or use a reference-to-video model |
| The camera moves when it shouldn't | Put "Static camera" or "Locked-off static camera" early in the prompt, or try a model that follows camera directions more closely |
| A model you use is going away | Read its Knowledge Center page for the replacement, for example Gemini Omni 1.1 Flash to 1.2 Flash |

## What's next

With a model chosen, [generate the video](https://learn.scenario.com/video/generate-video.md) with your first and last frames, or learn how to [prompt video](https://learn.scenario.com/methods/prompt-video.md) so each take moves the way you want.

Source: https://learn.scenario.com/video/video-models-guide