Ultra Selection — Intelligent AI Model Routing API
Intelligent model routing that analyzes your prompt and automatically selects the best AI model based on quality, speed, or cost — so you never have to guess.
How Ultra Selection Works
A three-step intelligent routing process that optimizes every generation request for quality, speed, or cost.
Prompt Analysis & Model Matching
Ultra Selection analyzes your prompt complexity, subject matter, and style requirements in real-time. It matches your request against our model performance database to identify the optimal model for your specific use case.
Strategy-Based Routing
Choose your priority: best-quality for maximum visual fidelity, fastest-response for speed-critical applications, or lowest-cost for budget-conscious batch processing. Ultra Selection routes accordingly.
Transparent Model Attribution
Every API response includes full metadata showing which model processed your request. You maintain complete visibility into routing decisions and can override selections at any time.
Ultra Selection vs. Manual Model Picking
See why teams choose Ultra Selection over manual model picking.
Performance at a Glance
Ultra Selection on WaveSpeed delivers intelligent routing at scale.
Examples
Young woman turning to smile at camera, breeze catching her scarf, soft bokeh background.
Dancer performing a graceful pirouette, flowing dress creating motion trails, spotlight.
Butterfly emerging from chrysalis in close-up, wings slowly unfurling, soft natural light.
Detective walking through foggy city streets, trench coat collar up, film noir atmosphere.
Integrate in Minutes
Production-ready SDKs for Python and JavaScript. REST API with full OpenAPI spec. Webhook support for async jobs.
- Zero-config model routing with strategy parameter
- Full model attribution in every response
- Python & JavaScript SDKs + REST API
Get Any Tool You Want
1000+ models across image, video, audio, and 3D — all through one API.
FAQ
Ultra Selection is an API feature that acts as a "Meta-Model." Instead of specifying a model ID (like flux-dev), you specify a strategy (like best-quality or fastest-response), and WaveSpeed handles the rest.
Yes. You can configure the selection logic based on three priorities: Quality (always pick the best visual fidelity), Speed (always pick the fastest inference), or Cost (always pick the most economical model).
Currently, Ultra Selection is optimized for Image Generation. Video support is in beta, routing between models like Wan 2.1 and Kling based on prompt duration and motion complexity.
Ultra Selection itself is free to use. You are billed based on the actual model that ends up processing your request. If the system routes to a cheaper model, you pay less.
Yes. The API response metadata always includes the model_id of the model that actually performed the generation, ensuring transparency.