This repository provides a community-driven platform for comparing and exploring Large Language Models (LLMs), offering detailed benchmark scores, provider pricing, and model specifications. It serves researchers, developers, and users seeking to understand and select LLMs based on performance and cost.

How It Works

The project aggregates data on hundreds of LLMs, including parameters, context window sizes, licensing, capabilities, pricing, and performance metrics. Benchmark results are standardized and sourced with verifiable links, undergoing community review to ensure data quality and accuracy.

Quick Start & Requirements

The primary interface is the interactive dashboard available at llm-stats.com. No specific installation or local setup is required to view the data.

Highlighted Details

• Comprehensive data on hundreds of LLMs.
• Includes model parameters, context window sizes, licensing, and capabilities.
• Features provider pricing and performance metrics (throughput, latency).
• Standardized benchmark results with verifiable source links.

Maintenance & Community

The project is community-driven, with contributions welcomed via pull requests following contribution guidelines. Discussions and community engagement are facilitated through a Discord server.

Licensing & Compatibility

The repository's license is not explicitly stated in the provided README.

Limitations & Caveats

While the project strives for accuracy, there is no guarantee that the data is 100% accurate. Some benchmark scores are marked as "N/A," indicating missing data for specific models and tests.