Best LLM for every budget, updated daily

This interactive dashboard compares every model in the Artificial Analysis Intelligence Index by plotting intelligence score against a blended API price, then surfaces the value frontier - models for which nothing cheaper is also smarter. Points on the chart are flagged as either on the value frontier or dominated, and defaults show only each model's best variant while hiding low scorers; filters let users expose every variant, set a minimum score, or label each point. The core use is a "best model for your budget" lookup that returns the highest-scoring model affordable at a given blended price per 1M tokens and its runner-up, so users can quickly see which models give the most capability per dollar.

Prices use a 3:1 input:output blend per 1M tokens (log scale) and exclude cached-input discounts, batch pricing and fast-mode pricing. The full table includes maker, intelligence, coding and math scores, blended/input/output $/1M, tokens/sec and time-to-first-token. The site also shows daily diffs (new, removed, re-scored or re-priced models). Scores can shift because the Intelligence Index is rebased between versions. Data comes from the Artificial Analysis API and is refreshed daily via GitHub Actions; code and raw data are available on GitHub for inspection.