Towards AITowards AIToneBench

Claude Fable 5 (max) vs GPT-5.6 Sol (high)

Should you write with Claude Fable 5 (max) or GPT-5.6 Sol (high)? On this page the only thing that changes is the model. Both wrote their own version of the scripts behind 10 of my What's AI videos, five drafts per script, and three judges from three model families scored every draft blind, with my finished script as the reference. Claude Fable 5 (max) leads overall, 89.7 to 87.6. Heads-up: GPT-5.6 Sol (high) shares its model line with one of the three judges. The panel spans three model families, so no model here is scored by its own family alone (how the panel works).

Claude Fable 5 (max)
#5 Elo 2392 · 89.7/100
GPT-5.6 Sol (high)
#33 Elo 2134 · 87.6/100
Cost / script
$2.522 vs $0.155
Human baseline
90.9 Claude Fable 5 (max) falls below it · GPT-5.6 Sol (high) falls below it

The verdict

I'd use Claude Fable 5 (max effort + 4.8 fallback) for the writing, but GPT-5.6 Sol (high) is a better deal than its #33 suggests next to Fable's #5. Fable's edge is YouTube craft and voice: three to four points ahead on both YouTube Best Practices and Tone & Voice Match. GPT-5.6 Sol (high) wins Visual Cue Quality, the lightest metric in the rubric, and it's fast, under a minute and a half per script against about nine, at around fifteen cents instead of two and a half dollars. For structured, cue-heavy scripts on a budget, that's a sensible pick.

Pick Claude Fable 5 (max effort + 4.8 fallback) when the voice and YouTube structure need to be right without a rewrite.
Pick GPT-5.6 Sol (high) for fast, inexpensive scripts with strong visual cues.

How to read these numbers

Overall vs Elo. Overall is the weighted rubric score. Elo compares each model's five-draft average with every other model's, script by script, and counts gaps inside run-to-run noise as draws, so it rewards winning often and can disagree with overall. My own scripts score 90.9 on the same rubric, not a literal 100. Right now 2 of 167 ranked configurations reach it. A model above that line writes at that level on this rubric. That's not the same as beating my script line for line.

Reading the Elo intervals. Each 95% range comes from resampling every model's recorded run scores on each task and re-ranking. Comparing these separate ranges does not test the Elo gap between two models: far apart is strong evidence, and overlap isn't a tie. The tasks stay fixed, so the ranges say nothing about new tasks (the long version).

Metric by metric: where the gap comes from

Blue is Claude Fable 5 (max), orange is GPT-5.6 Sol (high), and the solid bar takes the metric. The weight under each name is how much it counts toward the overall. Weighted by how much each metric counts, Tone & Voice Match does the most to put Claude Fable 5 (max) ahead: it leads there by 3.2 points. If you care about one thing, say voice or length, go straight to that row.

Tone & Voice Match19% weight
89.7
86.5
Writing Craft & Clarity13% weight
89.6
88.1
Substance, Accuracy & Value15% weight
88.6
88.2
Continuity & Emotion14% weight
89.1
86.1
YouTube Best Practices12% weight
90.5
86.8
Hook Strength10% weight
88.6
86.6
Length Adherence8% weight
93.2
89.3
Slop Score (EQ-Bench + ours)5% weight
93.4
92.8
Visual Cue Quality4% weight
85.4
89.4

Cost, speed, and consistency

Quality is half the decision. The other half is what each script costs you, how long you wait, and how often you get a bad draft. The lead has a price: Claude Fable 5 (max) costs about 16x as much per script as GPT-5.6 Sol (high).

Claude Fable 5 (max)GPT-5.6 Sol (high)
Overall / 10089.787.6
Writing Elo23922134
Score spread across drafts (± overall std)1.3601.330
Cost per script (USD)2.5220.155
Avg time per script (min)9.11.4
Open weightsNoNo

Green wins the row, and for spread, cost and time, lower wins. Cost is one script at recorded list prices, uncached and without judging, so your real bill will differ. Spread is how much the overall score moves across a model's drafts, and time is the average time to get one accepted script, retries included.

Want the script-by-script detail? Full scorecards: Claude Fable 5 (max) · GPT-5.6 Sol (high). The whole scoring pipeline is on the methodology page.

← All comparisons