Name: Google Veo 3.1 Fast – Text to Video | $0.1/call | API.market
Brand: google
SKU: api-google-veo-31-fast
Availability: InStock
Author: API Market

Pulkit

# 🎬 Veo 3.1 Fast — Video Generation API

## Overview

The **Veo 3.1 Fast** API lets you generate high-fidelity, cinematic videos from text prompts — with optional first and last frame image guidance for precise control over the scene.This API delivers fast, production-quality video generation with optional AI-generated audio.

Pricing is per second of generated video: **$0.10/sec without audio**, **$0.15/sec with audio**.

---

## Features

- **Text-to-Video Generation** — Describe any scene, action, or style in text and generate a cinematic video clip.
- **Image-Guided Generation** — Provide a first frame, a last frame, or both to precisely anchor your video's start and end.
- **AI Audio Generation** — Optionally generate synchronized audio alongside the video.
- **Flexible Duration** — Choose 4, 6, or 8 seconds of output video.
- **Aspect Ratio Control** — Landscape (16:9) or portrait (9:16).
- **Person Generation Control** — Allow or restrict generation of people in the output.
- **Seed Support** — Use a fixed seed for reproducible outputs.

---

## Use Cases

- **Cinematic Short Clips** — Generate dramatic, high-quality video sequences for film production, trailers, or storyboarding.
- **Social Media Content** — Produce scroll-stopping short-form video in portrait or landscape format.
- **Product Visualization** — Animate product shots with smooth motion and cinematic lighting.
- **Game & Metaverse Assets** — Rapidly prototype in-game cinematics, cutscenes, and environment flythroughs.
- **Advertising Campaigns** — Generate video ad creatives from a brief text description of the scenario.
- **Creative Storytelling** — Turn written narratives into visual video sequences with controlled pacing using first/last frame guidance.

---

## Visual Example: Image-Guided Video Generation

**First Frame (input image):**

![First Frame](https://blog.api.market/wp-content/uploads/2026/03/image-5.png)

**Last Frame (input image):**

![Last Frame](https://blog.api.market/wp-content/uploads/2026/03/image-6.png)

**Prompt:** `"A cinematic fantasy story scene. A group of medieval warriors walks through a smoky rocky valley toward a burning castle at sunset. The camera slowly pushes forward from behind the warriors, with drifting embers in the air, moving cloaks, glowing firelight, atmospheric smoke, dramatic lighting, realistic motion, epic cinematic composition, high detail."`

**Output:** A 6-second cinematic video with audio —
### Output Video
![fantasy story scene](https://blog.api.market/wp-content/uploads/2026/03/sample_0.mp4)

---

## Pricing

| Audio          | Price per Second |
|----------------|-----------------|
| Without Audio  | $0.10 / second  |
| With Audio     | $0.15 / second  |


- `GET /video/status/{id}` status-check requests are **free** and do not count toward usage.
- A minimum wallet balance of $1.00 is required to use the API.

---

## API Documentation

### Base URL

```
https://prod.api.market/api/v1/magicapi/veo-31-fast
```

### Authentication

Include your API key in every request header:

```
x-api-market-key: YOUR_API_MARKET_KEY
```

---

### Asynchronous Processing

> **Important:** This API processes requests asynchronously. You must poll for results.

The standard flow is:

1. Submit your request to `POST /video/run`.
2. Receive a `request_id` in the response.
3. Poll `GET /video/status/{id}` using the `request_id`.
4. Keep polling until `status` is `"succeeded"` or `"failed"`.

Implement your polling logic with:

- A **3–5 second** interval between checks (video generation typically takes 30–90 seconds).
- A **600-second (10-minute)** maximum timeout.
- Handling for `"queued"`, `"dispatched"`, `"processing"`, `"succeeded"`, `"failed"`, and `"cancelled"` statuses.

---

### Endpoints

---

#### 1. Submit Video Generation Request

**`POST /video/run`**

Submits a request to generate a video using Veo 3.1 Fast.

##### Request Headers

| Header             | Value              |
|--------------------|--------------------|
| `Content-Type`     | `application/json` |
| `x-api-market-key` | Your API key       |

##### Request Body Schema

```json
{
  "input": {
    "prompt": "string",
    "image": "string",
    "lastFrame": "string",
    "durationSeconds": 6,
    "aspectRatio": "16:9",
    "generateAudio": true,
    "negativePrompt": "string",
    "personGeneration": "allow_all",
    "seed": 42
  }
}
```

##### Request Body Parameters

| Parameter               | Type    | Required | Default        | Description |
|-------------------------|---------|----------|----------------|-------------|
| `input.prompt`          | string  | Yes      | —              | Text description of the video to generate (max 2000 characters). Describe scenes, actions, camera moves, and visual style. |
| `input.image`           | string  | No       | —              | URL of the first frame image for two-frame guided video generation. Supported formats: JPEG, PNG, WebP. Max 1 image. Note: image-to-video only supports 8-second duration. |
| `input.lastFrame`       | string  | No       | —              | URL of the last frame image for two-frame guided video generation. Supported formats: JPEG, PNG, WebP. Max 1 image. |
| `input.durationSeconds` | integer | No       | `6`            | Video length in seconds. Options: `4`, `6`, `8`. Note: image-guided generation only supports `8`. |
| `input.aspectRatio`     | string  | No       | `"16:9"`       | Aspect ratio of the generated video. Options: `"16:9"`, `"9:16"`. |
| `input.generateAudio`   | boolean | No       | `true`         | Whether to generate audio for the video. Affects pricing: `true` = $0.15/sec, `false` = $0.10/sec. |
| `input.negativePrompt`  | string  | No       | —              | Describe what you don't want to see in the video (max 500 characters). |
| `input.personGeneration`| string  | No       | `"allow_all"`  | Control whether people can appear. Options: `"allow_all"`, `"allow_adult"`, `"disallow"`. |
| `input.seed`            | integer | No       | —              | Random seed for reproducibility (range: 0–4294967295). Leave empty for random results. |

##### Request Body Examples

**Text-to-video (no audio):**

```json
{
  "input": {
    "prompt": "A cinematic drone shot flying low over a misty rainforest at dawn. Ancient trees tower overhead, shafts of golden light pierce the canopy, birds scatter as the camera glides through. Lush, atmospheric, photorealistic.",
    "durationSeconds": 6,
    "aspectRatio": "16:9",
    "generateAudio": false
  }
}
```

**Image-guided video with first and last frame:**

```json
{
  "input": {
    "prompt": "A cinematic fantasy story scene. A group of medieval warriors walks through a smoky rocky valley toward a burning castle at sunset. The camera slowly pushes forward from behind the warriors, with drifting embers in the air, moving cloaks, glowing firelight, atmospheric smoke, dramatic lighting, realistic motion, epic cinematic composition, high detail.",
    "image": "https://blog.api.market/wp-content/uploads/2026/03/image-5.png",
    "lastFrame": "https://blog.api.market/wp-content/uploads/2026/03/image-6.png",
    "durationSeconds": 6,
    "aspectRatio": "16:9",
    "generateAudio": true,
    "negativePrompt": "blurry, low quality, distorted faces, extra limbs, deformed hands, unrealistic motion, flickering, artifacts, cartoon, low detail, bad anatomy",
    "personGeneration": "allow_all",
    "seed": 42
  }
}
```

**Short portrait video with audio:**

```json
{
  "input": {
    "prompt": "Ocean waves crashing on a rocky shore at golden hour, dramatic spray, warm light, cinematic slow motion",
    "durationSeconds": 4,
    "aspectRatio": "9:16",
    "generateAudio": true,
    "negativePrompt": "cartoon, flat lighting, low quality"
  }
}
```

##### Response Body Schema

```json
{
  "request_id": "string",
  "model": "string",
  "status": "string",
  "created_at": 0,
  "updated_at": 0,
  "queued_at": 0
}
```

##### Response Body Parameters

| Field        | Type    | Description |
|--------------|---------|-------------|
| `request_id` | string  | Unique request identifier. Use this to poll `GET /video/status/{id}`. |
| `model`      | string  | Model used (`"veo-3.1-fast-generate-preview"`). |
| `status`     | string  | Current status. One of: `"queued"`, `"dispatched"`, `"processing"`, `"succeeded"`, `"failed"`, `"cancelled"`. |
| `created_at` | integer | Unix timestamp (seconds) when the request was created. |
| `updated_at` | integer | Unix timestamp (seconds) when the request was last updated. |
| `queued_at`  | integer | Unix timestamp (seconds) when the request was queued. |

##### Response Examples

**Queued (awaiting processing):**

```json
{
  "request_id": "fe57b39f-893f-4cd9-9757-7b47dd4ba628",
  "model": "veo-3.1-fast-generate-preview",
  "status": "dispatched",
  "created_at": 1773835627,
  "updated_at": 1773835627,
  "queued_at": 1773835627
}
```

##### cURL Example

```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/veo-31-fast/video/run" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "input": {
      "prompt": "A cinematic fantasy story scene. A group of medieval warriors walks through a smoky rocky valley toward a burning castle at sunset. The camera slowly pushes forward from behind the warriors, with drifting embers in the air, moving cloaks, glowing firelight, atmospheric smoke, dramatic lighting, realistic motion, epic cinematic composition, high detail.",
      "image": "https://blog.api.market/wp-content/uploads/2026/03/image-5.png",
      "lastFrame": "https://blog.api.market/wp-content/uploads/2026/03/image-6.png",
      "durationSeconds": 6,
      "aspectRatio": "16:9",
      "generateAudio": true,
      "negativePrompt": "blurry, low quality, distorted faces, extra limbs, deformed hands, unrealistic motion, flickering, artifacts, cartoon, low detail, bad anatomy",
      "personGeneration": "allow_all",
      "seed": 42
    }
  }'
```

---

#### 2. Check Video Generation Status

**`GET /video/status/{id}`**

Polls the status of a video generation request and retrieves the output URL when complete.

##### Request Headers

| Header             | Value        |
|--------------------|--------------|
| `x-api-market-key` | Your API key |

##### Path Parameters

| Parameter | Type   | Required | Description |
|-----------|--------|----------|-------------|
| `id`      | string | Yes      | The `request_id` returned from `POST /video/run`. |

##### Response Body Schema

```json
{
  "request_id": "string",
  "model": "string",
  "status": "string",
  "output": "string",
  "payload": {
    "prompt": "string",
    "durationSeconds": 0,
    "aspectRatio": "string",
    "generateAudio": true
  },
  "outcome": {
    "video_url": "string",
    "thumbnail_image_url": "string"
  },
  "created_at": 0,
  "updated_at": 0,
  "queued_at": 0
}
```

##### Response Body Parameters

| Field                          | Type    | Description |
|-------------------------------|---------|-------------|
| `request_id`                  | string  | Unique request identifier. |
| `model`                       | string  | Model used (`"veo-3.1-fast-generate-preview"`). |
| `status`                      | string  | Current status. One of: `"queued"`, `"dispatched"`, `"processing"`, `"succeeded"`, `"failed"`, `"cancelled"`. |
| `output`                      | string  | Direct URL to the generated video (MP4). Present when `status` is `"succeeded"`. |
| `payload`                     | object  | The original request parameters submitted. |
| `payload.prompt`              | string  | The text prompt used. |
| `payload.durationSeconds`     | integer | Duration of the video in seconds. |
| `payload.aspectRatio`         | string  | Aspect ratio used. |
| `payload.generateAudio`       | boolean | Whether audio was generated. |
| `outcome`                     | object  | Full generation result. Present when `status` is `"succeeded"`. |
| `outcome.video_url`           | string  | Public URL to download the generated video (MP4 format). |
| `outcome.thumbnail_image_url` | string  | URL of a JPEG thumbnail preview of the generated video. |
| `created_at`                  | integer | Unix timestamp (seconds) when the request was created. |
| `updated_at`                  | integer | Unix timestamp (seconds) when the request was last updated. |
| `queued_at`                   | integer | Unix timestamp (seconds) when the request was queued. |

##### Response Examples

**Succeeded:**

```json
{
  "request_id": "fe57b39f-893f-4cd9-9757-7b47dd4ba628",
  "model": "veo-3.1-fast-generate-preview",
  "status": "succeeded",
  "output": "https://storage.googleapis.com/gmi-video-assests-prod/user-assets/fb079f50-8bb7-47ac-962f-b2f54ac1a845/e7c3bf8d-3586-4fec-88de-c1da2f41d6fc/gmi-videogen/generated/fe57b39f-893f-4cd9-9757-7b47dd4ba628.mp4/1620516255316310256/sample_0.mp4",
  "payload": {
    "prompt": "A cinematic fantasy story scene...",
    "durationSeconds": 6,
    "aspectRatio": "16:9",
    "generateAudio": true,
    "image": "https://blog.api.market/wp-content/uploads/2026/03/image-5.png",
    "lastFrame": "https://blog.api.market/wp-content/uploads/2026/03/image-6.png",
    "negativePrompt": "blurry, low quality...",
    "personGeneration": "allow_all",
    "seed": 42
  },
  "outcome": {
    "video_url": "https://storage.googleapis.com/gmi-video-assests-prod/user-assets/fb079f50-8bb7-47ac-962f-b2f54ac1a845/e7c3bf8d-3586-4fec-88de-c1da2f41d6fc/gmi-videogen/generated/fe57b39f-893f-4cd9-9757-7b47dd4ba628.mp4/1620516255316310256/sample_0.mp4",
    "thumbnail_image_url": "https://storage.googleapis.com/gmi-video-assests-prod/user-assets/fb079f50-8bb7-47ac-962f-b2f54ac1a845/e7c3bf8d-3586-4fec-88de-c1da2f41d6fc/gmi-videogen/generated/thumbnail_image_fe57b39f-893f-4cd9-9757-7b47dd4ba628.jpg"
  },
  "created_at": 1773835627,
  "updated_at": 1773835690,
  "queued_at": 1773835627
}
```

**Still queued:**

```json
{
  "request_id": "fe57b39f-893f-4cd9-9757-7b47dd4ba628",
  "model": "veo-3.1-fast-generate-preview",
  "status": "queued",
  "created_at": 1773835627,
  "updated_at": 1773835627,
  "queued_at": 1773835627
}
```

##### cURL Example

```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/veo-31-fast/video/status/fe57b39f-893f-4cd9-9757-7b47dd4ba628" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

---

## Complete Integration Example

Here is a full end-to-end example in Python demonstrating how to submit a generation request and poll for the result:

```python
import requests
import time

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/veo-31-fast"
HEADERS = {
    "Content-Type": "application/json",
    "x-api-market-key": API_KEY,
}

# Step 1: Submit video generation request
response = requests.post(
    f"{BASE_URL}/video/run",
    headers=HEADERS,
    json={
        "input": {
            "prompt": "A cinematic fantasy story scene. A group of medieval warriors walks through a smoky rocky valley toward a burning castle at sunset. The camera slowly pushes forward from behind the warriors, with drifting embers in the air, moving cloaks, glowing firelight, atmospheric smoke, dramatic lighting, realistic motion, epic cinematic composition, high detail.",
            "image": "https://blog.api.market/wp-content/uploads/2026/03/image-5.png",
            "lastFrame": "https://blog.api.market/wp-content/uploads/2026/03/image-6.png",
            "durationSeconds": 6,
            "aspectRatio": "16:9",
            "generateAudio": True,
            "negativePrompt": "blurry, low quality, distorted faces, extra limbs",
            "personGeneration": "allow_all",
            "seed": 42,
        }
    },
)
response.raise_for_status()
request_id = response.json()["request_id"]
print(f"Request submitted. ID: {request_id}")

# Step 2: Poll for result
max_wait = 600  # 10 minutes
interval = 5   # seconds between polls
elapsed = 0

while elapsed < max_wait:
    time.sleep(interval)
    elapsed += interval

    status_response = requests.get(
        f"{BASE_URL}/video/status/{request_id}",
        headers=HEADERS,
    )
    status_response.raise_for_status()
    data = status_response.json()
    status = data.get("status")
    print(f"[{elapsed}s] Status: {status}")

    if status == "succeeded":
        video_url = data.get("output") or data.get("outcome", {}).get("video_url")
        thumbnail_url = data.get("outcome", {}).get("thumbnail_image_url")
        print(f"✅ Video ready!")
        print(f"   Video URL:     {video_url}")
        print(f"   Thumbnail URL: {thumbnail_url}")
        break
    elif status in ("failed", "cancelled"):
        print(f"❌ Generation {status}.")
        break
else:
    print("⏱️ Timeout: video generation took longer than expected.")
```

---

## Parameter Reference

### `durationSeconds`

| Value | Notes |
|-------|-------|
| `4`   | Shortest clip. Fastest generation. |
| `6`   | Default. Best balance of speed and content. |
| `8`   | Longest clip. Required when using `image` or `lastFrame` for image-guided generation. |

### `aspectRatio`

| Value   | Use Case |
|---------|----------|
| `"16:9"` | Landscape — YouTube, desktop, cinematic. Default. |
| `"9:16"` | Portrait — TikTok, Instagram Reels, mobile-first content. |

### `generateAudio`

| Value   | Price      | Notes |
|---------|------------|-------|
| `true`  | $0.15/sec  | AI-generated ambient audio synchronized to the video. |
| `false` | $0.10/sec  | Silent video output. Useful when you'll add custom audio in post. |

### `personGeneration`

| Value          | Description |
|----------------|-------------|
| `"allow_all"`  | Allow generation of all people, including minors. Default. |
| `"allow_adult"` | Allow generation of adults only. |
| `"disallow"`   | No people will appear in the generated video. |

---

## Error Handling

| HTTP Status | Meaning | Common Cause |
|-------------|---------|--------------|
| `400`       | Bad Request | Invalid parameter value (e.g., unsupported `durationSeconds`, `aspectRatio`). |
| `401`       | Unauthorized | Missing or invalid `x-api-market-key`. |
| `404`       | Not Found | Invalid `request_id` passed to the status endpoint. |
| `500`       | Internal Server Error | Unexpected server-side error. Retry with exponential backoff. |

When `status` is `"failed"`, the generation itself failed on the model side. Retry the request with a revised prompt or different parameters.

---

## Tips for Better Results

- **Be specific in your prompt** — describe camera movement (e.g., "slow push forward", "drone shot pulling back"), lighting, atmosphere, and subject behavior.
- **Use `negativePrompt`** — exclude artifacts like `"blurry, flickering, low quality, cartoon, distorted"` to improve output consistency.
- **First and last frame guidance** — providing both `image` and `lastFrame` gives the model strong anchors for scene composition and dramatically reduces unwanted motion.
- **Set a `seed`** — use a fixed seed during prompt iteration to isolate the effect of prompt changes without random variation.
- **`generateAudio: false` for faster iteration** — when testing prompts, skip audio to reduce cost and get results faster.


Generate high-fidelity, cinematic videos from text prompts. Features optional image guidance and AI-generated audio. From $0.1/call. Try it free on API.market.