LLM Leaderboard - Best AI Models Ranked by Benchmarks & Price

Based on our live composite scoring across benchmarks, pricing, speed, and capabilities, GPT-5.4 Pro currently leads the AI model leaderboard with a score of 94. Rankings are updated hourly using real-time data from 51+ providers.

We use a composite scoring system (0-100) that combines: capabilities (25%), pricing tier (25%), context window (15%), recency (15%), output capacity (10%), and versatility (10%). Data is sourced from OpenRouter API and refreshed hourly.

We track 331 AI models across 51 providers including OpenAI, Anthropic, Google, DeepSeek, Meta, and Mistral. This includes 306 coding models, 14 image generation models, and 10 video generation models.

Our coding category ranks 306 models by their performance on coding benchmarks like SWE-bench, HumanEval, and real-world coding tasks. The top coding models are evaluated hourly. Visit our coding leaderboard for the current top 10.