AI Model Benchmarks and Pricing Dataset 2026: Large Language Model Performance Survey
Description
A comprehensive survey of artificial intelligence language model performance and pricing economics as of 2026, maintained by BenchGecko (https://benchgecko.ai). This dataset covers benchmark evaluations across multiple dimensions including general knowledge (MMLU, MMLU-Pro), coding ability (HumanEval, SWE-bench Verified), mathematical reasoning (MATH, GSM8K, AIME), graduate-level science (GPQA Diamond), and instruction following (IFEval, AlpacaEval). Pricing data covers cross-provider API costs normalized to USD per million tokens. Resources and tools for working with this data: Model Rankings: https://benchgecko.ai/models Side-by-Side Comparison: https://benchgecko.ai/compare Cross-Provider Pricing: https://benchgecko.ai/pricing Free API: https://benchgecko.ai/api-docs AI Economy Dashboard: https://benchgecko.ai/economy Compute Supply Chain: https://benchgecko.ai/compute Mindshare Arena: https://benchgecko.ai/mindshare MCP Server Directory: https://benchgecko.ai/mcp Agent Directory: https://benchgecko.ai/agents Changelog: https://benchgecko.ai/changelog Methodology: Scores sourced from original technical reports and cross-verified using open-source evaluation frameworks. Pricing collected from official API documentation, updated within 48 hours of changes. Full methodology at https://benchgecko.ai/methodology