Thinking Levels · xAI
Last updated , from board data generated at 17:52 UTC.
This page compares Grok 4.7 at 3 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, Grok 4.7 ranks #28 of 87 models through its best setting. The other ranks on this page are configuration ranks: one setting's place among all 151 configurations.
Raising the effort lifts Grok 4.7, but you pay a lot for it. Going from low to high adds 52 Elo, so high would win about 57% of script comparisons against low, though the two ranges overlap. The bill is the catch: 4.7 times the cost per script, and 9.6 minutes of waiting instead of 1.7.
Almost all of that gain is length. At low, Grok 4.7 misses the target length badly, and medium fixes a good part of it, while voice and substance barely move and the craft score even slips a little. Going on to high adds almost nothing on length. So it comes down to whether you fix the length by hand anyway. If you do, low at $0.064 is the one I'd run.
If you want scripts that come back closer to the right length, medium is the setting on this ladder. It has the highest Writing Elo of the three, with a range that overlaps both of the others, and high only adds a little on cues for more money and a longer wait. At $0.215 a script and 6.2 minutes, though, I'd look elsewhere first: DeepSeek V4.1 Flash (max) ranks higher for $0.017.
Each point is one effort setting of Grok 4.7, 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 | Reasoning Tokens | Configuration Rank |
|---|---|---|---|---|---|---|---|
| low | 1757 ±51 | - | 79.0 | $0.064 | 1.7 min | 2,760 | #64 |
| medium | 1840 ±51 | +83 | 79.6 | $0.215 | 6.2 min | 27,687 | #57 |
| high | 1809 ±46 | -31 | 79.6 | $0.305 | 9.6 min | 42,549 | #58 |
The 95% Writing Elo ranges overlap: Grok 4.7 (low) runs from 1708 to 1810, and Grok 4.7 (high) from 1761 to 1853.
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 previous explicit setting. 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 xAI family page lists every xAI model side by side.