# llm.ing

> Aggregated LLM benchmarks, prices, and specs with per-score provenance.
> No bot walls: this site is meant to be readable by humans and machines alike.

## Key pages

- /leaderboard : sortable cross-benchmark leaderboard (intelligence, coding, agentic, Arena Elo, price, context)
- /models/{lab}/{id} : per-model page — scores with provenance and source links, pricing per 1M tokens, provider uptime/latency/throughput
- /benchmarks : directory of aggregated benchmarks with caveats and licensing
- /methodology : how aggregation, provenance labels, and model-identity mapping work
- /status : per-source data freshness

## Machine-readable data

- /api/v1/models.json : model registry with specs and pricing (models.dev MIT + OpenRouter)
- /api/v1/leaderboard.json : redistributable scores only (LMArena Elo, CC BY 4.0, attribution required)

## Attribution

Intelligence/coding/agentic indices displayed on the site come from Artificial Analysis
(https://artificialanalysis.ai) and are not included in the JSON API per their terms.
Arena Elo data © LMArena, CC BY 4.0.