GPT-5.6 Sol (ultra) vs Grok 4.6 (high)
Head-to-head on the Towards AI writing benchmark: same 10 real YouTube scripts, five runs each, scored blind by a three-family judge panel. GPT-5.6 Sol (ultra) leads overall, 88.3 to 88.1. Their confidence intervals overlap, so treat the order as close rather than settled.
The verdict
GPT-5.6 Sol takes this one, but the board calls it a coin flip. It ranks sixteenth at 2157.5 Elo against Grok's seventeenth at 2144.5, a gap of 13 points, and the confidence intervals overlap, so treat the ordering as provisional. The overall scores agree: 88.35 versus 88.14, a difference of 0.21. The metric splits are what separate them. Sol is far more disciplined on length, 94.84 to Grok's 88.26, and its visual cues are cleaner, 90.59 against 86.43. It also leads on substance accuracy, 90.21 to 87.95. Grok answers with openings and voice: hook strength 90.03 to Sol's 86.59, tone 88.75 to 86.92. So Sol writes the tidier, more accurate script and Grok the more engaging opening. Cost barely moves the needle: about twelve cents per script for Grok, measured exactly, versus an estimated fourteen cents for Sol. Neither is open weights, so this comes down to which failure mode you can live with.
Pick GPT-5.6 Sol (ultra) if you need scripts that hit their length targets, keep visual cues clean, and get facts right, and you can live with a flatter opening.
Pick Grok 4.6 (high) if hooks and voice matter more to you than length discipline, and a slightly cheaper, exactly measured cost is a welcome bonus.
Metric by metric
Blue bars: GPT-5.6 Sol (ultra). Orange bars: Grok 4.6 (high). Same 0–100 scale; the bold bar wins that metric.
Everything else that differs
| GPT-5.6 Sol (ultra) | Grok 4.6 (high) | |
|---|---|---|
| Overall / 100 | 88.3 | 88.1 |
| Writing Elo | 2158 | 2144 |
| Run-to-run spread (± overall std) | 1.850 | 1.430 |
| Cost per script (USD) | 0.144 | 0.115 |
| Avg latency (s) | 417.9 | 199.2 |
| Open weights | No | No |
Full scorecards: GPT-5.6 Sol (ultra) · Grok 4.6 (high). How scoring works: methodology.
← All comparisons