Thinking Levels · Thinking Machines
Inkling 2 Small: Writing Quality by Reasoning Effort
Last updated , from board data generated at 17:52 UTC.
This page compares Inkling 2 Small 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, Inkling 2 Small ranks #31 of 87 models through its best setting. The other ranks on this page are configuration ranks: one setting's place among all 151 configurations.
What More Effort Changes for Inkling 2 Small
On Inkling 2 Small, xhigh comes out a little ahead on Writing Elo and thinks a lot longer to get there. Tinker runs the model at high when the request sets no level, and xhigh, the level Thinking Machines used for its own benchmarks, lands 32 Elo above that run, #67 against #68. Their ranges overlap, and overlap alone doesn't say which setting writes better. Tinker doesn't report reasoning tokens on their own, but the total output per draft nearly doubles at xhigh, for scripts that are only a little longer. The wait goes up about as much, to 2.2 minutes a script instead of 1.2. It costs 1.6 times as much, which still comes to only $0.026 a script.
Recommended Setting
The scores show where that extra thinking goes: length. At xhigh, the scripts land much closer to their target length, and the visual cues tick up a little. Every other score stays within about half a point, and craft even slips a hair.
So I'd run it at xhigh. The Elo lead alone doesn't prove it writes better, but it hands you scripts closer to the length you asked for, and on a bill this small, paying 1.6 times as much for that is an easy call. Stay on high when you want lots of drafts back fast and you'll fix the length by hand anyway.
Every Setting on the Chart
Each point is one effort setting of Inkling 2 Small, joined from the lowest setting to the highest. Higher and further left is better: a higher score for less money or less time.
How to Use the Chart
- A hollow marker is the provider default: the level the provider applies when the request sets no effort level.
- The tabs choose the score on the vertical axis, X axis switches between cost and time, and Table lists the same numbers as rows.
- Select a point for its numbers and a link to its scorecard.
On a small screen, Table lists the same numbers as rows.
Both Effort Settings Compared
| Setting | Writing Elo | Step | Writing Score | Cost / Script | Time / Script | Configuration Rank |
|---|---|---|---|---|---|---|
| high (provider default) | 1696 ±55 | - | 78.1 | $0.016 | 1.2 min | #68 |
| xhigh | 1728 ±43 | +32 | 78.9 | $0.026 | 2.2 min | #67 |
The 95% Writing Elo ranges overlap: Inkling 2 Small (high) runs from 1639 to 1748, and Inkling 2 Small (xhigh) from 1684 to 1771.
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 Thinking Machines family page lists every Thinking Machines model side by side.