Thinking Levels · Anthropic
Last updated , from board data generated at 17:52 UTC.
This page compares Claude Opus 5 at 2 reasoning-effort settings, each tested on the same 10 scripts: the Writing Score (the 0 to 100 consensus of the judges and code), the Writing Elo (a rating from pairwise comparisons of those scores), and the cost and time of each setting. As a model, Claude Opus 5 ranks #6 of 87 models through its best setting. The other ranks on this page are configuration ranks: one setting's place among all 151 configurations.
Max effort gives Claude Opus 5 a lead at a fairly modest price. Max lands 108 Elo above the run with no effort flag (high), going from #20 to #16, and the two ranges just overlap, though overlap alone doesn't test their difference. Max costs 1.6 times as much, $0.736 a script, and takes about 5.5 minutes instead of 3.5.
On Claude Opus 5, I'd still run max. The lead is on the point estimate, but at that price it's a cheap bet. For anything new, though, I'd go one version up. Claude Opus 5.5 (high) ranks well above both settings at #8, costs $0.328 a script, which is less than this model with no flag, and comes back faster too. At high, the newer Opus writes better scripts for less money than either setting here.
Each point is one effort setting of Claude Opus 5, joined from the lowest setting to the highest. Higher and further left is better: a higher score for less money or less time.
On a small screen, Table lists the same numbers as rows.
| Setting | Writing Elo | Step | Writing Score | Cost / Script | Time / Script | Configuration Rank |
|---|---|---|---|---|---|---|
| high (provider default) | 2147 ±52 | - | 82.6 | $0.472 | 3.5 min | #20 |
| max | 2255 ±54 | +108 | 83.2 | $0.736 | 5.5 min | #16 |
The 95% Writing Elo ranges overlap: Claude Opus 5 (high) runs from 2097 to 2201, and Claude Opus 5 (max effort) from 2197 to 2305.
Writing Elo comes from comparing every pair of models on every script, which puts these settings on the same scale as the leaderboard. Step is the Writing Elo change from the provider default. Cost is one full script at API list prices, and time is the average for one accepted script, retries included. The thinking-levels page covers the other models, and the Anthropic family page lists every Anthropic model side by side.