AI Image, Video & Music Generation API

One REST API for 53+ production models — Kling 3.0, SeeDance 2.0, Seedream, GPT Image 2, Wan 2.7, and more — plus content moderation, music-genre classification, and format conversion. Transparent per-second credit pricing, charged only on success, with an exact cost estimate available before every request.

Get an API key Read the docs

Using Claude, ChatGPT, or Cursor? Connect without code via the BudgetPixel MCP server. In ComfyUI, WordPress or Cursor? Every model here ships as a plugin, including the ComfyUI node pack.

What you get access to

AI models available: 53+
Community creations published: 10K+
Royalty-free audio tracks: 1.8K+
Maximum output resolution: 4K
Total generations served platform-wide: 4.5M+

Two requests to your first video

1 · Create a job

curl -X POST https://api.budgetpixel.com/v1/videos/kling-v3.0-pro \
  -H "Authorization: Bearer $BUDGETPIXEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "a paper boat drifting across a puddle in gentle rain", "length_seconds": 5}'

# → {"id": "5ff73bb0-...", "status": "pending", "model": "kling-v3.0-pro"}

2 · Poll until it succeeds

curl https://api.budgetpixel.com/v1/videos/5ff73bb0-... \
  -H "Authorization: Bearer $BUDGETPIXEL_API_KEY"

# → {"status": "succeeded", "video_url": "https://..."}

Watch it in action

From API key to finished generation — the flow above, end to end.

Everything behind the studio, as an API

Video generation

Text-to-video, image-to-video with start/end frames, multi-reference, and video-to-video editing across the Kling, SeeDance, Wan, and HappyHorse families.

Image generation

Text-to-image, editing, and multi-reference composition — Seedream, FLUX, GPT Image 2, Nano Banana, Qwen, and more.

Music generation

Full songs and instrumentals from a prompt or lyrics, plus a music-genre classification endpoint for any uploaded track.

Content moderation

The same NSFW rating and CSAM detection classifiers the platform runs on itself, exposed as synchronous endpoints for your own content.

Transparent cost estimation

POST /v1/cost prices any request before you run it, computed by the exact functions that bill — the estimate can't drift from the invoice.

Utilities included

File uploads for media inputs, format conversion for images, video, and audio, and publishing straight to the BudgetPixel community feed.

Every model, one API

The same models, called from your code

One key and one credit balance across image, video, music and sound effects. Switching model is a parameter, not an integration.

Video generation

Seedance 2.5

Cinematic video up to 30 seconds. Text-to-video, image-to-video with a start and end frame, or reference-driven video mixing image, video and audio clips. 480p, 720p or 1080p.

curl -X POST https://api.budgetpixel.com/v1/videos/seedance-2.5 \
  -H "Authorization: Bearer bpx_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a lighthouse in a storm, slow push-in",
    "resolution": "720p",
    "duration_seconds": 5
  }'

About Seedance 2.5 API reference

Video generation

Wan 3.0

All-in-one video: text, image and reference-to-video in a single endpoint, with every input medium free — you are billed on output seconds only.

curl -X POST https://api.budgetpixel.com/v1/videos/wan-3.0-video \
  -H "Authorization: Bearer bpx_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "neon-lit alley, rain, camera drifts left",
    "resolution": "480p",
    "duration_seconds": 5
  }'

About Wan 3.0 API reference

Image generation

Seedream 5.0 Pro

Flagship image quality with the strongest identity preservation in the catalogue — the default when an edit has to keep a face or product consistent.

curl -X POST https://api.budgetpixel.com/v1/images/seedream-5.0-pro \
  -H "Authorization: Bearer bpx_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "portrait of a lighthouse keeper, film grain",
    "aspect_ratio": "4:3",
    "resolution": "2K"
  }'

About Seedream 5.0 Pro API reference

Image generation

GPT Image 2.5 Sunburst

The pick when the image has to contain legible text — posters, UI mockups, packaging — and for precision edits that change only what you asked for.

curl -X POST https://api.budgetpixel.com/v1/images/gpt-image-2.5-sunburst \
  -H "Authorization: Bearer bpx_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "poster reading OPEN LATE in bold type",
    "quality": "high",
    "resolution": "2K"
  }'

About GPT Image 2.5 Sunburst API reference

Browse the full model catalogue — prices and per-model input limits are also available at runtime from GET /v1/models.

Models and pricing

Prices are in platform credits at the standard metered rate — the same numbers the billing code computes. Estimate any exact request with POST /v1/cost.

Video models

Model	Endpoint	Pricing	Modes	Links
Kling v2.6 Pro	POST /v1/videos/kling-v2.6-pro	150 credits/sec	text-to-video, image-to-video	Docs
Kling 3.0 Standard	POST /v1/videos/kling-v3.0-standard	85 credits/sec	text-to-video, image-to-video	Docs · Examples
Kling 3.0 Pro	POST /v1/videos/kling-v3.0-pro	115 credits/sec	text-to-video, image-to-video	Docs · Examples
Kling 3.0 4K	POST /v1/videos/kling-v3.0-4k	450 credits/sec	text-to-video, image-to-video	Docs · Examples
Kling 3.0 Turbo	POST /v1/videos/kling-3.0-turbo	120–140 credits/sec	text-to-video, image-to-video	Docs · Examples
Kling V3 Omni	POST /v1/videos/kling-v3-omni-video	85–450 credits/sec	text-to-video, image-to-video, reference-to-video, video-to-video	Docs · Examples
MiniMax H3	POST /v1/videos/minimax-h3	160 credits/sec	text-to-video, image-to-video, reference-to-video	Docs · Examples
SeeDance 1.5 Pro	POST /v1/videos/seedance-1.5-pro	40–120 credits/sec	text-to-video, image-to-video	Docs · Examples
SeeDance 2.5	POST /v1/videos/seedance-2.5	150–750 credits/sec	text-to-video, image-to-video, reference-to-video	Docs · Examples
SeeDance 2.0	POST /v1/videos/seedance-2.0	100–1250 credits/sec	text-to-video, image-to-video, reference-to-video, video-to-video	Docs · Examples
SeeDance 2.0 Fast	POST /v1/videos/seedance-2.0-fast	80–160 credits/sec	text-to-video, image-to-video, reference-to-video, video-to-video	Docs · Examples
SeeDance 2.0 Mini	POST /v1/videos/seedance-2.0-mini	60–120 credits/sec	text-to-video, image-to-video, reference-to-video, video-to-video	Docs · Examples
Wan 3.0	POST /v1/videos/wan-3.0-video	60–240 credits/sec	text-to-video, image-to-video, reference-to-video	Docs · Examples
Wan 3.0 Prime	POST /v1/videos/wan-3.0-video-prime	85–340 credits/sec	text-to-video, image-to-video, reference-to-video	Docs · Examples
Wan 2.7	POST /v1/videos/wan-2.7-video	100–150 credits/sec	text-to-video, image-to-video, reference-to-video, video-to-video	Docs · Examples
HappyHorse 1.1	POST /v1/videos/happyhorse-1.1	170–200 credits/sec	text-to-video, image-to-video, reference-to-video	Docs · Examples
P-Video	POST /v1/videos/p-video	40–60 credits/sec	text-to-video, image-to-video	Docs

Image models

Model	Endpoint	Pricing	Modes	Links
SeeDream 4.5	POST /v1/images/seedream-4.5	45 credits/image	text-to-image, image editing / references	Docs · Examples
FLUX 2 Pro	POST /v1/images/flux-2-pro	25 credits/image	text-to-image, image editing / references	Docs · Examples
SeeDream 5.0 Lite	POST /v1/images/seedream-5.0-lite	40 credits/image	text-to-image, image editing / references	Docs · Examples
SeeDream 5.0 Pro	POST /v1/images/seedream-5.0-pro	55 credits/image	text-to-image, image editing / references	Docs · Examples
FLUX 2 Klein	POST /v1/images/flux-2-klein	10 credits/image	text-to-image, image editing / references	Docs · Examples
FLUX 2 Dev	POST /v1/images/flux-2-dev	25 credits/image	text-to-image, image editing / references	Docs · Examples
FLUX 1.1 Pro	POST /v1/images/flux-1.1-pro	40 credits/image	text-to-image, image editing / references	Docs
Qwen-Image	POST /v1/images/qwen-image	15 credits/image	text-to-image	Docs
Qwen Image 2.0	POST /v1/images/qwen-image-2.0	45 credits/image	text-to-image, image editing / references	Docs · Examples
Qwen Image 2.0 Pro	POST /v1/images/qwen-image-2.0-pro	85 credits/image	text-to-image, image editing / references	Docs · Examples
Qwen Image 3.0	POST /v1/images/qwen-image-3.0	40 credits/image	text-to-image, image editing / references	Docs · Examples
Qwen Image 3.0 Pro	POST /v1/images/qwen-image-3.0-pro	45 credits/image	text-to-image, image editing / references	Docs · Examples
Wan 2.7	POST /v1/images/wan-2.7	35 credits/image	text-to-image, image editing / references	Docs · Examples
Wan 2.7 Pro	POST /v1/images/wan-2.7-pro	80 credits/image	text-to-image, image editing / references	Docs · Examples
Z-Image-Turbo	POST /v1/images/z-image-turbo	15 credits/image	text-to-image	Docs · Examples
P-Image	POST /v1/images/p-image	10 credits/image	text-to-image	Docs
P-Image Edit	POST /v1/images/p-image-edit	15 credits/image	text-to-image, image editing / references	Docs
Nano Banana Pro	POST /v1/images/nano-banana-pro	120 credits/image	text-to-image, image editing / references	Docs · Examples
Nano Banana 2	POST /v1/images/nano-banana-2	65 credits/image	text-to-image, image editing / references	Docs · Examples
Nano Banana 2 Lite	POST /v1/images/nano-banana-2-lite	40 credits/image	text-to-image, image editing / references	Docs · Examples
GPT-Image-2	POST /v1/images/gpt-image-2	60 credits/image	text-to-image, image editing / references	Docs · Examples
GPT Image 2.5 Sunburst	POST /v1/images/gpt-image-2.5-sunburst	40 credits/image	text-to-image, image editing / references	Docs · Examples
GPT Image 2.5 Flare	POST /v1/images/gpt-image-2.5-flare	40 credits/image	text-to-image, image editing / references	Docs · Examples
Grok Imagine	POST /v1/images/grok-imagine-image	35 credits/image	text-to-image, image editing / references	Docs · Examples
Grok Imagine 2.0	POST /v1/images/grok-imagine-image-2	50 credits/image	text-to-image, image editing / references	Docs · Examples
Muse Image 1.0	POST /v1/images/muse-image-1.0	12 credits/image	text-to-image, image editing / references	Docs · Examples
Kling v3	POST /v1/images/kling-v3	35 credits/image	text-to-image, image editing / references	Docs · Examples
Kling v3 Omni	POST /v1/images/kling-v3-omni	35 credits/image	text-to-image, image editing / references	Docs · Examples

Music models

Model	Endpoint	Pricing	Modes	Links
Music 3.0	POST /v1/audios/music-3.0	200 credits/track	music generation	Docs · Examples
Music 2.6	POST /v1/audios/music-2.6	200 credits/track	music generation	Docs
Lyria 3	POST /v1/audios/lyria-3	100 credits/track	music generation	Docs · Examples
Mureka V9	POST /v1/audios/mureka-v9	60 credits/track	music generation	Docs · Examples
Sonilo Music	POST /v1/audios/sonilo-music	4 credits/sec	music generation	Docs · Examples
Sonilo Video Music	POST /v1/audios/sonilo-video-music	15 credits/sec	video-to-music	Docs · Examples

Sound effect models

Model	Endpoint	Pricing	Modes	Links
Sonilo SFX	POST /v1/audios/sonilo-sfx	5 credits/sec	text-to-sound-effect	Docs · Examples
Sonilo Video SFX	POST /v1/audios/sonilo-video-sfx	15 credits/sec	video-to-sound-effect	Docs · Examples

Start in three steps

1 · Get a key — subscribe to any paid plan, then mint a key in the developer console. Your key spends the same credit balance your plan includes.
2 · Send a request — media inputs accept public URLs, uploads via POST /v1/uploads, or base64. The OpenAPI-generated docs cover every parameter.
3 · Poll and ship — jobs survive restarts, failed generations are never billed, and results are delivered as signed URLs.

See plans and credits

Frequently asked questions

What is the BudgetPixel API?▾

A REST API (https://api.budgetpixel.com/v1) that gives developers programmatic access to the 53+ AI image, video, and music models running on BudgetPixel, plus content moderation, music-genre classification, format conversion, and cost estimation endpoints. One key, one integration pattern, OpenAPI-documented at docs.budgetpixel.com.

How do I get an API key?▾

Create keys in the developer console at budgetpixel.com/developers. API access is included with every paid plan — no separate API contract, and your key shares the credit balance your plan already includes.

How does API pricing work?▾

Usage is metered in credits at the standard rate — for example video models bill per second of output, at the rates listed on this page. Your plan's model discounts and active promotions apply to API charges exactly as they do on the web, and Pro/Ultra plans' daily free-models allowance (100/300 free generations per day, pooled across qualified image models) covers API requests too. You are only charged when a generation succeeds; failed jobs cost nothing. POST /v1/cost returns the exact price of any request before you run it.

Which video models are available through the API?▾

Kling v2.6 Pro, Kling 3.0 Standard, Kling 3.0 Pro, Kling 3.0 4K, Kling 3.0 Turbo, Kling V3 Omni, and more — each supporting some combination of text-to-video, image-to-video (start and end frames), reference-images-to-video, and video-to-video editing. See the tables on this page for per-model modes and per-second pricing.

Which image models are available through the API?▾

SeeDream 4.5, FLUX 2 Pro, SeeDream 5.0 Lite, SeeDream 5.0 Pro, FLUX 2 Klein, FLUX 2 Dev, and more — covering text-to-image, instruction-based editing, and multi-reference composition up to 4K output.

Is generation synchronous or asynchronous?▾

Image, video, and music generation are asynchronous: the create call returns a job id immediately and you poll its status endpoint until it succeeds (typically seconds for images, 1–5 minutes for video). Moderation, classification, and conversion endpoints are synchronous.

Can I moderate my own content with the BudgetPixel API?▾

Yes — POST /v1/moderations/nsfw rates text, images, or video on a 4-level NSFW scale, and POST /v1/moderations/csam runs the same child-safety classifier the platform uses internally. There is also POST /v1/classifications/music-genre for audio genre detection.

Get an API key