Model Reference

Side-by-side context window sizes for every major model, sortable and filterable.

Tokens-per-second benchmarks for every major hosted LLM.

Which models support vision, audio, tool use, and structured JSON — at a glance.

Per-million-token prices for every major model — sortable by input, output, and cache rate.

Find the right open weights model for your VRAM budget and task.

TPM, RPM, and tier limits for OpenAI, Anthropic, Google, and friends.