GPT-5.6 Sol (ultra) vs Grok 4.7 (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, 88.0 to 87.8. Their per-model Elo intervals overlap on this board.

GPT-5.6 Sol (ultra)

#25 Elo 2187 · 88.0/100

Grok 4.7 (high)

#26 Elo 2186 · 87.8/100

Cost / script

$0.363 vs $0.311

Human baseline

90.9 GPT-5.6 Sol (ultra) falls below it · Grok 4.7 (high) falls below it

The verdict

This one is closer than the ranks suggest. GPT-5.6 Sol (ultra) sits at #25 with 2186.8 Elo and 88.05 overall; Grok 4.7 (high) is at #26 with 2186.4 Elo and 87.79 overall. Their 95% Elo intervals overlap (2146.7 to 2223.2 against 2130.8 to 2232.5), so do not read much into the exact order. Where GPT-5.6 Sol (ultra) pulls ahead: Length Adherence (91.90 vs 74.12) and Visual Cue Quality (90.11 vs 88.85). Grok 4.7 (high) still wins on Tone & Voice Match (89.77 vs 86.19) and Continuity & Emotion (88.01 vs 86.03), so it is not a clean sweep. The trade is price: Grok 4.7 (high) costs about $0.311 per article against $0.363 for GPT-5.6 Sol (ultra), so you pay for the higher rank. Both are closed. Default to GPT-5.6 Sol (ultra). Reach for Grok 4.7 (high) when tone, continuity, and the lower price matter more to you than the headline rank.

Pick GPT-5.6 Sol (ultra) for the stronger board result, length adherence, and visual cues.
Pick Grok 4.7 (high) for tone, continuity, and the lower price.

Reading the Elo intervals. Each range estimates uncertainty around one model's Elo. We repeatedly sample from every model's recorded run scores on each task and recalculate the full ranking. Comparing these separate ranges does not test the Elo gap between two models. The tasks stay fixed, so the ranges do not measure how the ranking would change on new tasks.

Metric by metric

Blue bars: GPT-5.6 Sol (ultra). Orange bars: Grok 4.7 (high). Same 0–100 scale; the bold bar wins that metric.

Tone & Voice Match19% weight

86.2

89.8

Writing Craft & Clarity13% weight

88.1

87.9

Substance, Accuracy & Value15% weight

89.0

88.6

Continuity & Emotion14% weight

86.0

88.0

YouTube Best Practices12% weight

87.7

89.4

Hook Strength10% weight

86.8

88.1

Length Adherence8% weight

91.9

74.1

Slop Score (EQ-Bench + ours)5% weight

93.4

93.3

Visual Cue Quality4% weight

90.1

88.8

Everything else that differs

	GPT-5.6 Sol (ultra)	Grok 4.7 (high)
Overall / 100	88.0	87.8
Writing Elo	2187	2186
Run-to-run spread (± overall std)	1.230	2.670
Cost per script (USD)	0.363	0.311
Avg latency (s)	195.5	519.0
Open weights	No	No

Full scorecards: GPT-5.6 Sol (ultra) · Grok 4.7 (high). How scoring works: methodology.

← All comparisons