> ## Documentation Index > Fetch the complete documentation index at: https://testweb2-docs-openapi.test765.com/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > Five API is a video generation REST API. Base URL: https://api.fiveapi.ai. Authenticate with the header Authorization: Bearer ; read the key from the FIVEAPI_API_KEY environment variable and never hard-code it. > Every generation is an asynchronous task: POST /v1/{brand}/{version}/{task} returns a queued task, then poll GET /v1/tasks/{task_id} or receive a signed webhook, and download output.video.url before expires_at (7 days). > The OpenAPI spec is at https://docs.fiveapi.ai/api-reference/openapi.yaml. An installable agent skill: npx skills add https://docs.fiveapi.ai # Text to video > Generate a video from a written description, with optional synchronized audio. Text to video turns a prompt into a video. It is available on three models: | Model | Endpoint | Resolutions | Duration | Audio | | - | - | - | - | - | | Video Fast 1.0 | `POST /v1/video-fast/v1.0/text-to-video` | 480p, 768p | 3–15 s | — | | Wan 3.0 Spark | `POST /v1/wan/v3.0-spark/text-to-video` | 480p, 720p, 1080p | 2–30 s | yes | | Wan 3.0 Prime Spark | `POST /v1/wan/v3.0-prime-spark/text-to-video` | 480p, 720p, 1080p | 2–30 s | yes | ## Create a task ```bash cURL theme={null} curl https://api.fiveapi.ai/v1/wan/v3.0-spark/text-to-video \ -H "Authorization: Bearer $FIVEAPI_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: campaign-42-shot-1" \ -d '{ "prompt": "a lighthouse on a cliff during a storm, waves crash below, slow aerial orbit, cinematic", "negative_prompt": "text, watermark, blurry", "duration": 6, "resolution": "1080p", "aspect_ratio": "16:9", "generate_audio": true, "webhook_url": "https://hooks.example.com/fiveapi", "metadata": { "campaign": "42", "shot": 1 } }' ``` ```python Python theme={null} import os import requests task = requests.post( "https://api.fiveapi.ai/v1/wan/v3.0-spark/text-to-video", headers={ "Authorization": f"Bearer {os.environ['FIVEAPI_API_KEY']}", "Idempotency-Key": "campaign-42-shot-1", }, json={ "prompt": "a lighthouse on a cliff during a storm, waves crash below, slow aerial orbit, cinematic", "negative_prompt": "text, watermark, blurry", "duration": 6, "resolution": "1080p", "aspect_ratio": "16:9", "generate_audio": True, "webhook_url": "https://hooks.example.com/fiveapi", "metadata": {"campaign": "42", "shot": 1}, }, timeout=30, ).json() print(task["id"], task["status"]) ``` ```javascript Node.js theme={null} const res = await fetch("https://api.fiveapi.ai/v1/wan/v3.0-spark/text-to-video", { method: "POST", headers: { Authorization: `Bearer ${process.env.FIVEAPI_API_KEY}`, "Content-Type": "application/json", "Idempotency-Key": "campaign-42-shot-1", }, body: JSON.stringify({ prompt: "a lighthouse on a cliff during a storm, waves crash below, slow aerial orbit, cinematic", negative_prompt: "text, watermark, blurry", duration: 6, resolution: "1080p", aspect_ratio: "16:9", generate_audio: true, webhook_url: "https://hooks.example.com/fiveapi", metadata: { campaign: "42", shot: 1 }, }), }); const task = await res.json(); console.log(task.id, task.status); ``` The task comes back `queued` with `credits.estimated` held. Get the video from the `task.completed` [webhook](https://testweb2-docs-openapi.test765.com/concepts/webhooks.md) or by polling [Get a task](https://testweb2-docs-openapi.test765.com/api-reference/tasks/get-task.md). ## Parameters | Field | Required | Description | | - | - | - | | `prompt` | yes | What to generate, up to 2,000 characters. | | `duration` | yes | Length in seconds, within the model's range. | | `resolution` | yes | One of the model's resolutions. | | `aspect_ratio` | no | `9:16`, `16:9`, `1:1`, `3:4` or `4:3`. Default: `9:16` on Video Fast 1.0, `16:9` on Wan 3.0. | | `negative_prompt` | no | What to keep out of the video, up to 2,000 characters. | | `generate_audio` | no | Wan 3.0 only. `true` adds synchronized audio. Default `false`. | | `prompt_expansion` | no | `none`, `basic` or `advanced` (default). With `basic` or `advanced` the prompt is rewritten into a more detailed description; the result is in `output.actual_prompt`. | | `seed` | no | Fixed seed for more reproducible results. | | `webhook_url`, `metadata` | no | See [Webhooks](https://testweb2-docs-openapi.test765.com/concepts/webhooks.md) and [Tasks](https://testweb2-docs-openapi.test765.com/concepts/tasks.md). | The exact ranges of each model are on its API reference page and in `GET /v1/models`. ## Write better prompts A good prompt describes the shot the way a director would brief a camera crew: | Element | Example | | - | - | | Subject | "an elderly fisherman with a white beard" | | Action | "mends a net, then looks up at the sea" | | Setting and light | "on a wooden pier at dawn, golden backlight, light fog" | | Camera | "slow dolly-in, shallow depth of field" | | Style | "cinematic, 35 mm film look" | | Sound (with `generate_audio`) | "gulls cry, waves lap against the pier" | Tips: * Describe one continuous shot. Several unrelated scenes in one prompt tend to blur together. * Name the camera movement explicitly: "static shot", "slow pan left", "aerial orbit". * Use `negative_prompt` for things you keep seeing and don't want, such as text, watermarks or extra limbs. * Leave `prompt_expansion` at `advanced` for short prompts. Switch to `none` when you have written a detailed prompt and want it used exactly. * Iterate at 480p and a short duration, then render the final version at a higher resolution with the same prompt and `seed`. ## Choose the aspect ratio | Ratio | Typical use | | - | - | | `9:16` | Vertical video for Reels, TikTok and Shorts | | `16:9` | Landscape video for YouTube, web pages and TV | | `1:1` | Square posts | | `4:3`, `3:4` | Classic photo framing, product shots | This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.