NewSEEDANCE 2.5 NOW AVAILABLE · TRY IT IN THE PLAYGROUND →Get Started

One pool. Every model.

Generate anything.

Every top AI image and video model in one place. One account, one balance, no setup per provider.

Prompt
— — —
Start Creating See All Models

// Gallery

Forged on FluxPool

A sprawling medieval marketplace viewed from above, dozens of merchant stalls with distinct awnings, crowds bartering, geese underfoot. Flux 2 Pro Forged on FluxPool
A girl on her bicycle, cycling along a sunny beach with a bunch of colorful helium balloons trailing behind the bicycle. Seedream 5.0 Forged on FluxPool
A close-up of the inside of a beehive. Flux 2 Pro Forged on FluxPool
A chaotic day in an 1800s New York Stock Exchange with brokers crowding and hustling to buy and sell shares. Seedream 5.0 Forged on FluxPool
A parkour climber taking a selfie from the very top of a Manhattan skyscraper Qwen Image 2.0 Pro Forged on FluxPool
A ballerina doing a dance on top of a moving train. Flux 2 Pro Forged on FluxPool
A viking standing at the head of a ship as it navigates a tumultuous ocean. Seedream 5.0 Forged on FluxPool
A weathered fisherman mending a net on a fog-covered dock at dawn, shot on 85mm with soft golden backlight and shallow depth of field. Flux 2 Pro Forged on FluxPool
Chalkboard menu outside a Parisian café with elegant handwritten script listing 'Croissant, Café au Lait, Tarte aux Pommes.' Qwen Image Max Forged on FluxPool
A sprawling medieval marketplace viewed from above, dozens of merchant stalls with distinct awnings, crowds bartering, geese underfoot. Flux 2 Pro Forged on FluxPool
A girl on her bicycle, cycling along a sunny beach with a bunch of colorful helium balloons trailing behind the bicycle. Seedream 5.0 Forged on FluxPool
A close-up of the inside of a beehive. Flux 2 Pro Forged on FluxPool
A chaotic day in an 1800s New York Stock Exchange with brokers crowding and hustling to buy and sell shares. Seedream 5.0 Forged on FluxPool
A parkour climber taking a selfie from the very top of a Manhattan skyscraper Qwen Image 2.0 Pro Forged on FluxPool
A ballerina doing a dance on top of a moving train. Flux 2 Pro Forged on FluxPool
A viking standing at the head of a ship as it navigates a tumultuous ocean. Seedream 5.0 Forged on FluxPool
A weathered fisherman mending a net on a fog-covered dock at dawn, shot on 85mm with soft golden backlight and shallow depth of field. Flux 2 Pro Forged on FluxPool
Chalkboard menu outside a Parisian café with elegant handwritten script listing 'Croissant, Café au Lait, Tarte aux Pommes.' Qwen Image Max Forged on FluxPool

// Model Catalog

Featured models

image

Dola-Seedream 5.0 Pro

Flagship image model with striking detail and strong prompt adherence. Up to 10 reference images, and a Layers mode that splits an image into separate layers.

2560×1440 ~8s 9 CR
Try seedream-5.0 →
Example generation

Example generations

image new

Dola-Seedream 5.0 Flash

Seedream 5.0 at a fraction of Pro’s price and faster, with slightly less detail. Same sizes, references and Layers mode.

2560×1440 Faster than Pro 2 CR
Try seedream-5.0-flash →
Example generations coming soon

Example generations

image new

Flux 2 Max

The top Flux model, with the best quality and prompt following. When a prompt asks for something current, it looks it up on the web before drawing.

2048×2048 Slower than Pro 7 CR
Try flux-2-max →
Example generations coming soon

Example generations

image POPULAR

Flux 2 Pro

Photorealism, sharp detail and excellent prompt adherence. Paint a region of an image to regenerate just that part.

2048×2048 ~12s 3 CR
Try flux-2-pro →
Example generation

Example generations

image new

Flux 2 Klein

The fastest, lowest-priced Flux model. Same sizes and references as Pro. Good for quick drafts and high volume.

2048×2048 Faster than Pro 2 CR
Try flux-2-klein →
Example generations coming soon

Example generations

image

Qwen Image 2.0 Pro

Excellent Asian aesthetics. Strong text rendering. Multilingual prompts.

2048×2048 ~10s 8 CR
Try qwen-image-2.0-pro →
Example generation

Example generations

image

Qwen Image Max

Maximum quality Qwen image generation. Strong composition and detail.

1664×928 ~12s 8 CR
Try qwen-image-max →
Example generation

Example generations

video POPULAR

Seedance 2.5

Up to 30s single-pass clips with native audio and up to 30 image, 10 video and 10 audio references. Reference, Extend, Edit and First & Last modes. Draft at 480p, then render the final at 1080p.

1080p (final) 180–600s 50 CR
Try seedance-2.5 →

Example generations

video

Seedance 2.0

Multi-shot video up to 15s at up to 1080p, with sound and reference media. Reference and First & Last modes.

1080p 90–300s 34 CR
Try seedance-v2 →

Example generations

video new

Seedance 2.0 Fast

Seedance 2.0 at a lower price, at 480p or 720p. Same modes, references and sound as 2.0.

720p 90–300s 28 CR
Try seedance-v2-fast →
Example generations coming soon

Example generations

video new

Seedance 2.0 Mini

The lowest-priced Seedance, at 480p or 720p. Same modes, references and sound as 2.0.

720p 90–300s 17 CR
Try seedance-v2-mini →
Example generations coming soon

Example generations

video new

Wan 3.0 Prime

Wan 3.0 rendering faster, for about 40% more. 2–30s clips up to 1080p with native sound, and the same modes and references as Wan 3.0.

1080p ~60s (short clip) 33 CR
Try wan3.0-video-prime →
Example generations coming soon

Example generations

video new

Wan 3.0

2–30s clips at 480p, 720p or 1080p with native sound. Reference, First & Last, Edit and Extend modes. References can be images, video, audio, a document or a web page.

1080p 60–180s 25 CR
Try wan3.0-video →
Example generations coming soon

Example generations

video POPULAR

Wan Video 2.7

Text-to-video with multi-shot narrative, at 720p or 1080p.

1080p 60–300s 49 CR
Try wan2.7-t2v →

Example generations

video

Wan Video 2.7 · Reference

Keeps a subject’s identity from 1–5 reference images while the prompt drives motion and camera. 720p or 1080p, up to 10s.

1080p 60–300s 49 CR
Try wan2.7-r2v →
Example generations coming soon

Example generations

video

Runway Gen4.5

Runway’s flagship video model. Cinematic motion and coherence, up to 10s.

720p 120–300s 58 CR
Try runway-gen4.5 →

Example generations

Explore all models

// Developer

Drop in. Ship.

A single OpenAI-compatible endpoint. Change the base URL and your API key. Images, video, and edits — same schema, one credit balance.

Terminal — zsh 
curl https://api.fluxpool.ai/v1/images/generations \
  -H "Authorization: Bearer fp_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-2-pro",
    "prompt": "A dragon on a neon-lit Tokyo rooftop, 8K",
    "size": "1024x1024"
  }'

// Infrastructure

Engineered for velocity

One API. Every model.

One key, one balance, and the full pool of image and video models.

seedream-5.0seedream-5.0-flashflux-2-maxflux-2-proflux-2-kleinqwen-image-2.0-proqwen-image-maxseedance-2.5seedance-v2seedance-v2-fastseedance-v2-miniwan3.0-video-primewan3.0-videowan2.7-t2vwan2.7-r2vrunway-gen4.5

Fast inference

Pooled capacity across providers keeps queues short.

4× faster median
vs single-provider queues

Built for scale

From your first request to millions per day.

Requests / day · auto-scaling pool (illustrative)

Secure by default

Your prompts and outputs stay yours.

  • TLS 1.3 in transit
  • AES-256 at rest
  • Scoped API keys
  • Instant revocation
  • No training on customer data
  • Private by default

Start creating today

1-month Trial plan. No credit card required.

Start Creating See All Models