Leaderboards

Composite of every quality benchmark, normalised per benchmark and averaged per category. Requires results on at least five benchmarks across two categories.

BenchLeader Index divided by the blended price per million tokens (3 input : 1 output). Higher means more quality for the money.

Median output throughput measured across healthy providers over the last 30 minutes of live traffic (OpenRouter).

Median seconds until the answer starts, including any reasoning phase (Artificial Analysis), or time to first streamed token where that is all we have.

First-party list price where published, blended at 3 input tokens per output token.

BenchLeader Index divided by what it cost Artificial Analysis to run its whole Intelligence Index on that configuration. Unlike price per token this counts the tokens a model actually burns, so a model that thinks at length scores lower than its sticker price suggests.

What Artificial Analysis actually paid to run its whole Intelligence Index on each configuration: a fixed batch of work, billed at list price. The spread between models is far wider than their per-token prices, because verbose reasoning multiplies the token count.

Median seconds from request to the last token, reasoning included (Artificial Analysis). Time to first answer tells you when output starts; this tells you when it ends.

Maximum context length in tokens as listed by the provider.