Hub · Model Selection

Picking the right model

Meridian gives you access to dozens of models across multiple providers. This section helps you navigate the landscape — from choosing your first model to building sophisticated routing topologies that balance quality, speed, and cost.

Which model should I use?A decision framework for picking the right model based on task complexity, latency budget, and cost tolerance.Model routingHow Meridian's intelligent router selects the optimal model per request using quality signals and fallback chains.Provider routingConfigure provider priority, region affinity, and failover policies so your traffic always lands where it should.Models overviewThe complete catalogue of supported models across OpenAI, Anthropic, Google, Meta, Mistral, and open-weight runtimes.Models endpointProgrammatic access to model metadata, capabilities, pricing, and deprecation schedules via the REST API.Anti‑patternsCommon mistakes when selecting and routing models — over‑specifying, ignoring token limits, and provider lock‑in.Token estimationEstimate input and output tokens before you send a request so you can budget cost and stay within context windows.

Getting started

New to model selection? Start with Which model should I use? for a guided walkthrough, then explore model routing to automate decisions at scale.