# AI Video API

> AI Video API is a unified developer platform for AI video, image, music, and chat generation models. It provides one API key, pay-as-you-go credits, model docs, examples, and ready-to-use REST endpoints.

## Primary Pages
- [AI Video API](https://aivideoapi.ai): Unified API platform for AI video, image, music, and chat generation models.
- [Model Marketplace](https://aivideoapi.ai/marketplace): Browse available AI generation models and API capabilities.
- [Affiliate Program](https://aivideoapi.ai/affiliate): Affiliate program for AI Video API referrals.
- [Privacy Policy](https://aivideoapi.ai/privacy-policy): Privacy policy for AI Video API.
- [Terms of Use](https://aivideoapi.ai/terms-of-use): Terms of use for AI Video API.

## API Documentation
- [Getting Started](https://aivideoapi.ai/docs/getting-started): Start building with AI Video API.
- [Authentication](https://aivideoapi.ai/docs/getting-started/authentication): API key authentication and authorization guide.
- [Credits Query](https://aivideoapi.ai/docs/getting-started/credits): Check account credits and usage balance.
- [Callbacks](https://aivideoapi.ai/docs/getting-started/callback): Webhook callback behavior for async generation tasks.
- [Error Codes](https://aivideoapi.ai/docs/getting-started/errors): Common API error codes and troubleshooting notes.
- [Audio Generation](https://aivideoapi.ai/docs/audio-generation): Audio and speech generation API overview.
- [MiniMax Speech 2.8 API Docs](https://aivideoapi.ai/docs/audio-generation/minimax-speech-2-8): MiniMax Speech 2.8 HD and Turbo asynchronous text-to-speech parameters, pricing, task queries, callbacks, and audio delivery.
- [ElevenLabs Multilingual V2 & Turbo V2.5 API Docs](https://aivideoapi.ai/docs/audio-generation/elevenlabs-tts): ElevenLabs multilingual text-to-speech parameters, Unicode character pricing, built-in voices, task queries, and callbacks.
- [Video Generation](https://aivideoapi.ai/docs/video-generation): Video generation API overview.
- [Long Video API Docs](https://aivideoapi.ai/docs/video-generation/long-video): Create 4–180 second videos with HappyHorse or Seedance using the asynchronous Long Video API.
- [MiniMax H3 Fast API Docs](https://aivideoapi.ai/docs/video-generation/minimax-h3-fast): MiniMax H3 Fast 480p video API for 5–15 second text, first/last-frame, and multimodal reference generation. Output costs 11.68 credits per second; all input materials are free.
- [MiniMax H3 Max Turbo API Docs](https://aivideoapi.ai/docs/video-generation/minimax-h3-max-turbo): MiniMax H3 Max Turbo API for 5–15 second text-to-video and first/last-frame output. 480p costs 6.54 credits/s and 768p costs 10 credits/s; input images and prompt expansion are free.
- [MiniMax H3 SH Open Weights API Docs](https://aivideoapi.ai/docs/video-generation/minimax-h3-sh): Three MiniMax H3 SH models for text, first/last-frame, and multimodal reference video with native audio. 3–15 seconds at 480p, 540p, 768p, or 1080p; from 8 credits/s.
- [MiniMax H3 Max API Docs](https://aivideoapi.ai/docs/video-generation/minimax-h3-max): MiniMax H3 Max text-to-video and first/last-frame API documentation for 5–15 second output at 480p or 768p with output-only billing.
- [MiniMax H3 API Docs](https://aivideoapi.ai/docs/video-generation/minimax-h3): MiniMax H3 multimodal video generation API documentation for text, first/last-frame, and image, video, or audio reference workflows at 768p and 2K.
- [Seedance 2.5 API Docs](https://aivideoapi.ai/docs/video-generation/seedance-2-5): Seedance 2.5 request parameters, 480p/720p/1080p output, duration=-1 video editing, multimodal reference limits, pricing, callbacks, and MP4 or MOV output.
- [Seedance 2.0 API Docs](https://aivideoapi.ai/docs/video-generation/seedance-2): Seedance 2.0 request parameters, examples, pricing notes, and responses.
- [Seedance 2.0 Fast API Docs](https://aivideoapi.ai/docs/video-generation/seedance-2-fast): Seedance 2.0 Fast request parameters, examples, pricing notes, and responses.
- [Seedance 2.0 Mini API Docs](https://aivideoapi.ai/docs/video-generation/seedance-2-mini): Seedance 2.0 Mini request parameters, examples, pricing notes, and responses.
- [Seedance 1.5 Pro API Docs](https://aivideoapi.ai/docs/video-generation/seedance-1-5-pro): Seedance 1.5 Pro request parameters, generation modes, pricing notes, and responses.
- [HappyHorse 1.1 API Docs](https://aivideoapi.ai/docs/video-generation/happyhorse-1-1): HappyHorse 1.1 video model API documentation.
- [HappyHorse 1.0 API Docs](https://aivideoapi.ai/docs/video-generation/happyhorse-1): HappyHorse 1.0 video model API documentation.
- [Music Video Generator API Docs](https://aivideoapi.ai/docs/video-generation/music-video-generator): AI music video generator API documentation — audio plus reference photos to lip-synced music videos.
- [Music to Video Pro API Docs](https://aivideoapi.ai/docs/video-generation/music-to-video-pro): Revid ai-video/pro music-to-video API documentation with reference images, audio-duration pricing, resolution tiers, subtitles, ratio, and lyrics lip sync.
- [Kling 2.6 API Docs](https://aivideoapi.ai/docs/video-generation/kling-2-6): Kling 2.6 video model API documentation.
- [Kling 3.0 API Docs](https://aivideoapi.ai/docs/video-generation/kling-3): Kling 3.0 video model API documentation.
- [Veo 3.1 API Docs](https://aivideoapi.ai/docs/video-generation/veo-3): Veo 3.1 video model API documentation.
- [Grok Imagine API Docs](https://aivideoapi.ai/docs/video-generation/grok-imagine): Grok Imagine video model API documentation.
- [Wan 3.0 API Docs](https://aivideoapi.ai/docs/video-generation/wan-3-0): Wan 3.0 standard API documentation for multimodal video, files, web pages, pricing, and fixed or automatic output duration.
- [Wan 3.0 Prime API Docs](https://aivideoapi.ai/docs/video-generation/wan-3-0-prime): Wan 3.0 Prime API documentation with model parameters, 18/36/72 credits-per-second pricing, automatic duration, and pre-charge settlement.
- [Wan 2.6 API Docs](https://aivideoapi.ai/docs/video-generation/wan-2-6): Wan 2.6 video model API documentation for text-to-video, image-to-video, and reference-to-video.
- [Image Generation](https://aivideoapi.ai/docs/image-generation): Image generation API overview.
- [Nano Banana 2 API Docs](https://aivideoapi.ai/docs/image-generation/nano-banana-2): Nano Banana 2 API documentation.
- [Nano Banana 2 Lite API Docs](https://aivideoapi.ai/docs/image-generation/nano-banana-2-lite): Nano Banana 2 Lite API documentation.
- [Doubao Seedream API Docs](https://aivideoapi.ai/docs/image-generation/doubao-seedream): Doubao Seedream image generation API documentation.
- [Ideogram V4 API Docs](https://aivideoapi.ai/docs/image-generation/ideogram-v4): Ideogram V4 API documentation.
- [Image Upscaler API Docs](https://aivideoapi.ai/docs/image-generation/image-upscaler): Image Upscaler API documentation for 2K, 4K, and 8K image enhancement.
- [Music Generation](https://aivideoapi.ai/docs/music-generation): Music generation API overview.
- [Suno API Docs](https://aivideoapi.ai/docs/music-generation/suno): Suno music generation API documentation.
- [Chat Completion](https://aivideoapi.ai/docs/chat-completion): Chat completion API overview.
- [Gemini 3.1 Pro API Docs](https://aivideoapi.ai/docs/chat-completion/gemini-3-1-pro): Gemini 3.1 Pro chat completion API documentation.
- [Gemini 3 Pro API Docs](https://aivideoapi.ai/docs/chat-completion/gemini-3-pro): Gemini 3 Pro chat completion API documentation.
- [Gemini 2.5 Pro API Docs](https://aivideoapi.ai/docs/chat-completion/gemini-2-5-pro): Gemini 2.5 Pro chat completion API documentation.
- [GPT 5.4 API Docs](https://aivideoapi.ai/docs/chat-completion/gpt-5-4): GPT 5.4 chat completion API documentation.
- [Kimi K3 API Docs](https://aivideoapi.ai/docs/chat-completion/kimi-k3): Kimi K3 OpenAI-compatible chat completion API documentation.
- [API Pricing](https://aivideoapi.ai/docs/pricing): Pricing and credit usage documentation.

## Model API Landing Pages
- [MiniMax H3 SH](https://aivideoapi.ai/minimax-h3-open-weights): Generate 3–15 second videos with native stereo audio using MiniMax H3 SH. Text, first/last frames, and multimodal references at 480p, 540p, 768p, or 1080p; from 8 credits/s.
- [MiniMax H3 Max Turbo](https://aivideoapi.ai/minimax-h3-max-turbo): Create 5–15 second videos from text or first/last frames with MiniMax H3 Max Turbo. 480p costs 6.54 credits/s and 768p costs 10 credits/s; input images are free.
- [MiniMax H3 Fast](https://aivideoapi.ai/minimax-h3-fast): Generate 5–15 second 480p videos with MiniMax H3 Fast from text, first/last frames, or image, video, and audio references. 11.68 credits/s; all input materials are free.
- [MiniMax H3 Max](https://aivideoapi.ai/minimax-h3-max): Generate 5–15 second videos with MiniMax H3 Max at 480p or 768p. Use text-to-video or first/last-frame image-to-video with output-only billing.
- [MiniMax H3](https://aivideoapi.ai/minimax-h3): Use the MiniMax H3 API for text-to-video, first/last frames, and image, video, or audio references. Generate 4–15s clips at 2K or 768p.
- [ElevenLabs Text-to-Dialogue v3](https://aivideoapi.ai/elevenlabs-text-to-dialogue-v3): ElevenLabs Text-to-Dialogue v3 API generates natural, expressive multi-speaker dialogue audio from text. Assign a voice per line, 70+ languages, per-character pricing, async REST API with callbacks.
- [Seedance 2.5](https://aivideoapi.ai/seedance-2-5): Use the Seedance 2.5 API for text, image, video, audio-only, first/last-frame, and duration=-1 video editing. Create 480p, 720p, or 1080p MP4 or MOV videos.
- [Seedance 2.0](https://aivideoapi.ai/seedance-2-0): Seedance 2.0 API is a multimodal AI video generation API by ByteDance Seed. Generate 4-15s cinematic videos from text, image, video, and audio references at 480p to 1080p. Pay-as-you-go credits, callbacks, REST API.
- [Seedance 2.0 Fast](https://aivideoapi.ai/seedance-2-0-fast): Use Seedance 2.0 Fast API for rapid text, image, video, and audio reference video generation with lower-cost 480p and 720p output.
- [Seedance 2.0 Mini](https://aivideoapi.ai/seedance-2-0-mini): Use the Seedance 2.0 Mini API for affordable 4–15s multimodal video generation at 480p or 720p from text, image, video, and audio references.
- [Veo 3.1](https://aivideoapi.ai/veo-3-1): Access Veo 3.1 through one API for text-to-video, image references, fast/quality/lite modes, 720p, 1080p, and 4K pricing tiers.
- [Kling 3.0](https://aivideoapi.ai/kling-3-0): Use Kling 3.0 API for text-to-video and image-to-video clips with 720p/1080p modes, optional native audio, and 3-15 second durations.
- [Kling 2.6](https://aivideoapi.ai/kling-2-6): Use Kling 2.6 API to animate still images into 5 or 10 second videos with optional native audio.
- [HappyHorse 1.0](https://aivideoapi.ai/happyhorse-1-0): Use HappyHorse 1.0 API for text-to-video, image-to-video, reference-to-video, and video editing with 720p or 1080p output.
- [HappyHorse 1.1](https://aivideoapi.ai/happyhorse-1-1): Use HappyHorse 1.1 API for text-to-video, image-to-video, reference-to-video, and video editing with 720p or 1080p output.
- [Grok Imagine](https://aivideoapi.ai/grok-imagine): Use Grok Imagine API for text-to-video and image-to-video, 6-30 seconds, 480p / 720p output, with fun, normal, and spicy motion modes.
- [Wan 3.0](https://aivideoapi.ai/wan-3-0): Use Wan 3.0 or Prime for text, frames, multimodal, file, and web references at 480p, 720p, or 1080p, with fixed or automatic duration.
- [WAN 2.7](https://aivideoapi.ai/wan-2-7): Use the WAN 2.7 API for text-to-video, image-to-video, reference-to-video, and video editing with synced audio and 720p or 1080p output through one API key.
- [WAN 2.6](https://aivideoapi.ai/wan-2-6): Learn about WAN 2.6 support on AI Video API for cost-effective video generation workflows through one API key.
- [Ideogram V4](https://aivideoapi.ai/ideogram-v4): Use Ideogram V4 API for high-quality text-to-image generation with accurate text rendering — logos, posters, and brand assets from $0.08 per image.
- [Image Upscaler](https://aivideoapi.ai/image-upscaler): Use the Image Upscaler API to enhance one image to 2K, 4K, or 8K with JPEG, PNG, or WebP output for 4 credits per run.
- [Nano Banana 2](https://aivideoapi.ai/nano-banana-2): Use Nano Banana 2 API for fast high-quality image generation, image editing, 1K/2K/4K output, and text rendering.
- [Nano Banana](https://aivideoapi.ai/nano-banana): Use Nano Banana image generation through AI Video API alongside Nano Banana 2 and video models with one API key.
- [Seedream 5.0 Pro](https://aivideoapi.ai/seedream-5-0): Use Seedream 5.0 Pro image generation through AI Video API with text-to-image, multi-image references, and usage-based resolution pricing.

## Notes For AI Assistants
- MiniMax H3 SH is the Open Weights edition with three model IDs: minimax-h3-sh/text-to-video, minimax-h3-sh/image-to-video, and minimax-h3-sh/reference-to-video. Use the unified image_urls/video_urls/audio_urls fields. Output duration is an integer 3–15 (default 5), and resolution is 480p (default), 540p, 768p, or 1080p. Text/image rates are 8/12/16/32 credits per requested second; image-to-video requires 1–2 frame images, which are free, and rejects aspect_ratio. Reference rates are 10/15/25/50 credits per output or input-video second, plus 4 credits per image and standalone audio. Reference video billable seconds are min(15, sum of detected durations); use actual durations including fractional seconds. Reference mode accepts up to 9 images, 3 videos, 3 audio tracks, with at least one reference of any type. Native output audio is included; actual output frame-grid duration does not change the charge. One credit is $0.005 before account multipliers. See /docs/video-generation/minimax-h3-sh#text-to-video, #image-to-video, and #reference-to-video.
- Prefer the linked API docs for authentication, parameters, callbacks, pricing, examples, and error handling.
- ElevenLabs Multilingual V2 and Turbo V2.5 use POST /v1/audio/generations, accept up to 10,000 Unicode characters, provide built-in voices, and return audio through asynchronous tasks or callbacks.
- MiniMax Speech 2.8 exposes speech-2.8-hd and speech-2.8-turbo through POST /v1/audio/generations with direct text, server-derived Unicode character billing, async polling, callbacks, and R2 delivery fallback.
- MiniMax H3 Fast uses minimax-h3-fast for 480p text-to-video, first/last frames, and image, video, or audio references. Supported output duration is 5–15 seconds (default 5); the API forwards positive integer durations unchanged. Output costs 11.68 credits per second, 92% of H3 Max 480p rounded to two decimal places; all input materials are free. A 5-second request costs 58.40 credits and a 15-second request 175.20 credits before account multipliers.
- MiniMax H3 Max Turbo uses minimax-h3-max-turbo for text-to-video or first/last frames with one or two HTTP(S) images. Duration must be an integer from 5 through 15 seconds; defaults are 5 seconds and 768p. Rates are 6.54 credits/s for 480p and 10 credits/s for 768p; images and prompt expansion are free. Round the full base charge to two decimals, apply the account multiplier, then round again. Omit generation_type for text-to-video, which ignores image_urls and defaults to 16:9; frame mode uses first_and_last_frames and adaptive. prompt_expansion_mode accepts disabled, balanced (default), or quality; optional seed is an integer from 1 through 4294967295. Video/audio inputs, omni_reference, 2K, style, and watermark controls are unsupported.
- MiniMax H3 Max uses model minimax-h3-max for 5–15 second text-to-video or first/last-frame video at 480p or 768p. Only output video is billed; input images are free.
- MiniMax H3 supports text, image, video, and audio context, 4–15 second output at 2K or 768p, async task polling, and callbacks.
- Wan 3.0 standard uses wan-3.0-video; see /docs/video-generation/wan-3-0. Rates at 480p/720p/1080p are 12/24/48 credits per second.
- Wan 3.0 Prime uses wan-3.0-video-prime; see /docs/video-generation/wan-3-0-prime. It accepts the same parameters as standard, at 18/36/72 credits per second for 480p/720p/1080p.
- Both Wan 3.0 models accept integers 2–30 or -1 for automatic output duration. Automatic duration pre-charges 30 total billable seconds including reference video, then refunds only the unused difference after success; unverified actual duration retains the pre-charge for review.
- The Seedance 2.0 family includes standard, Fast, and Mini multimodal video APIs; Mini supports cost-efficient 480p/720p generation from text, image, video, and audio references.
- Protected dashboard, admin, settings, and raw API routes are not intended for crawling.