Artificial Analysis
Independent AI model benchmarks for intelligence, speed, pricing, context, and modalities.
Models & Infrastructure
Freemium4.5
§ Recipe
Production AI infrastructure.
A production AI stack is less about the flashiest model and more about control: routing, evals, fallback paths, open-model options, and cost visibility. This stack keeps you flexible while the model market keeps moving.
TeamSerious scaleOpen source
Model benchmark desk
01Start with independent quality, latency, and pricing data before you standardize on a model. It keeps model choice tied to current evidence instead of vendor claims.
Model routing
02OpenRouter
One API and routing layer for hundreds of AI models across many providers.
LLM Providers & APIsFreemium4.4Put a router in front of your app so switching models is a config decision, not a rewrite. It gives you fallback paths, price comparison, and faster evaluation cycles.
Open-model discovery
03Hugging Face
The central hub for AI models, datasets, Spaces, libraries, and open-source ML collaboration.
Models & InfrastructureFreemium4.8Use the hub for model cards, datasets, Spaces, and community signal. It is where you find candidates before they enter your production path.
Prototype model APIs
04Replicate
Run open and community AI models from a web playground or API.
Models & InfrastructurePaid4.4Turn model demos into API calls quickly, especially for image, video, and audio experiments. It is the fastest way to learn whether a model belongs in the product.
Serverless GPU jobs
05Modal
Serverless AI infrastructure for running code, jobs, containers, and GPUs from Python.
Models & InfrastructureFreemium4.5Run Python jobs, containers, scheduled work, and GPU endpoints without building cloud plumbing first. It is ideal for batch inference and AI backend work.
Production inference
06Baseten
Production AI inference platform for deploying, optimizing, and scaling models.
Models & InfrastructureEnterprise4.5When a model becomes a product dependency, Baseten handles serving, scaling, observability, and enterprise deployment patterns around that endpoint.