# Data

> Compare models across pricing, context length, benchmarks, throughput, and tool-calling accuracy. Data refreshes from public benchmarks and live perf telemetry.

**Web page**: https://anyrouter.dev/data

## Sections

- [Models](https://anyrouter.dev/data/models.md) — Overall model ranking by capability, throughput, and price/performance ratio.
- [Benchmark Scores](https://anyrouter.dev/data/benchmark.md) — Public benchmark results: Coding, GPQA Diamond, HLE, GDPval, τ²-Bench, and more.
- [Context Length](https://anyrouter.dev/data/context-length.md) — Models sorted by maximum context window size.
- [Speed](https://anyrouter.dev/data/speed.md) — Output tokens per second from tracked production telemetry over the last 30 days.
- [Tool Calling](https://anyrouter.dev/data/tool-calling.md) — Function-calling reliability scores across models that support tool use.
- [Apps](https://anyrouter.dev/data/apps.md) — Public apps and agents ranked by token volume.
- [Compare](https://anyrouter.dev/data/compare.md) — Side-by-side comparison of models across pricing, context length, benchmarks, and throughput.

## All models

See [all models](https://anyrouter.dev/models.md) for the full catalogue.
