Model Comparison
GPT-5.6 Terra vs Claude Opus 4.8
Too close to call: the best configurations, GPT-5.6 Terra (ultra) and Claude Opus 4.8 (max effort), are 61 Elo apart, and their 95% ranges overlap. Among 78 models, each ranked by its best configuration, GPT-5.6 Terra is #13 and Claude Opus 4.8 is #15. Cost per script: $0.275 for GPT-5.6 Terra (ultra) and $1.23 for Claude Opus 4.8 (max effort).
Last updated , from board data generated at 01:13 UTC.
Picks and Verdict
PickGPT-5.6 Terra
GPT-5.6 Terra (medium)
Pick GPT-5.6 Terra (medium) for most scripts at $0.062, about a twentieth of the price of Opus at max effort.
PickClaude Opus 4.8
Claude Opus 4.8 (max effort)
Pick Claude Opus 4.8 (max effort) when the structure and the length have to be right, since it leads both Terra settings on YouTube Best Practices and holds the length 7.5 points closer than Terra at medium, at $1.23 a script.
More thinking does about twice as much for Claude Opus 4.8 as for GPT-5.6 Terra, and I'd still run Terra. Opus gains 145 Elo from high to max effort, with separate 95% ranges. Terra gains 76 from medium to ultra, the two ranges overlap, and ultra costs more than four times as much. So GPT-5.6 Terra (medium) at $0.062 is the setting I'd compare everything to. By best setting Terra is the #13 model of 78 and Opus the #15 model, yet the range of Claude Opus 4.8 (max effort) overlaps both Terra settings, so the board can't separate them. It also costs $1.23, about twenty times Terra at medium, and takes seven times as long. Against Claude Opus 4.8 (high), Terra at medium leads by 130 Elo with separate ranges, at about a third of the price. Opus at max effort scores better on YouTube Best Practices and hits the length more closely. Terra at ultra leads on substance by over four points, and Terra at medium even edges Opus on tone. With results this close, medium is the Terra I'd run, and I'd spend the savings on editing.
Disclosure: GPT-5.6 Terra and Claude Opus 4.8 each come from the same lab as one of the judges (GPT-6.1 Sol and Claude Opus 5.5). Neither is scored by its own lab alone (how the panel works). The judge analysis measures how much each judge favors its own lab.
How to Read the Writing Score, Writing Elo and 95% Ranges
Writing Score and Writing Elo. The Writing Score (0 to 100) is the weighted average of the nine metric scores: seven from the judges, Length Adherence from code and the Slop Score half from each. The Writing Elo is a rating from pairwise comparisons: on each script, each configuration's average across its drafts is compared with every other configuration's, and gaps within the draft-to-draft variation count as draws. Elo rewards winning often, so the two can put configurations in a different order.
Elo ranges. Each 95% range shows how far a configuration's Elo could move if the drafts had come out slightly differently. It comes from resampling every configuration's recorded draft scores on each reference script and ranking again. Comparing two separate ranges does not test the Elo gap between them: ranges far apart are strong evidence for the order, and overlapping ranges do not mean a tie. The reference scripts stay fixed, so the ranges say nothing about new scripts. The methodology page has the full description.
All Thinking Levels of Both Models
Every ranked configuration of GPT-5.6 Terra and Claude Opus 4.8, scored on the same 10 scripts by the same four judges.
- Each row is one configuration: the model at one thinking level. The dot is its Writing Elo, a rating from pairwise comparisons of the scores on each script, and the bar is its 95% range: how far that Elo could move if the drafts had come out slightly differently.
- When two bars overlap, the board cannot order those two configurations. Bars far apart are strong evidence for the order.
- The table under the chart lists the same configurations in the same order and adds the Writing Score (the 0 to 100 consensus of the judges and code), cost, time and configuration rank (the place among all configurations on the board).
Shaded rows are the configurations chosen for the metric comparison below.
| Configuration | Writing Elo | 95% Range | Writing Score | Cost / Script | Time / Script | Configuration Rank |
|---|---|---|---|---|---|---|
| GPT-5.6 Terra (ultra) | 2074 | 2035 to 2113 | 81.4 | $0.275 | 8.0 min | #25 |
| Claude Opus 4.8 (max effort) | 2013 | 1977 to 2047 | 80.7 | $1.23 | 9.8 min | #30 |
| GPT-5.6 Terra (medium) | 1998 | 1952 to 2041 | 80.7 | $0.062 | 1.4 min | #33 |
| Claude Opus 4.8 (high) | 1868 | 1827 to 1905 | 79.3 | $0.193 | 1.4 min | #44 |
Cost is one script at the API list price of the route used, uncached and without judging. Time is the average for one accepted script, retries included.
Computed Findings Across Thinking Levels
Each finding is computed from the configurations above. Where two 95% ranges overlap, the board cannot order those two configurations.
- Best Configurations Head to Head. GPT-5.6 Terra (ultra) at Elo 2074 (range 2035 to 2113) and Claude Opus 4.8 (max effort) at Elo 2013 (range 1977 to 2047) have overlapping 95% ranges, so the board cannot separate the two best configurations.
- Lowest Thinking Levels Head to Head. At their lowest thinking levels, GPT-5.6 Terra (medium) at Elo 1998 and $0.062 per script leads Claude Opus 4.8 (high) at Elo 1868 and $0.193 per script by 130 Elo. Their 95% ranges do not overlap.
- Cheapest Configuration Overlapping the Top One. GPT-5.6 Terra (medium), at $0.062 per script, is the cheapest configuration whose 95% range overlaps that of GPT-5.6 Terra (ultra), at $0.275 per script: 1952 to 2041 against 2035 to 2113, so the board cannot separate them. GPT-5.6 Terra (ultra) costs about 4.4 times as much per script.
- Cheapest GPT-5.6 Terra Configuration Above Claude Opus 4.8's Best. The cheapest GPT-5.6 Terra configuration with a higher Writing Elo than Claude Opus 4.8 (max effort), at Elo 2013 and $1.23 per script, is GPT-5.6 Terra (ultra), at Elo 2074 and $0.275 per script. Their 95% ranges overlap, so the board cannot separate them. It costs less than the configuration it beats.
- Lowest to Highest Thinking Level. GPT-5.6 Terra moves +76 Elo from medium to ultra, at about 4.4 times the cost per script. Claude Opus 4.8 moves +145 Elo from high to max effort, at about 6.4 times the cost per script.
Two Chosen Configurations Compared
Pick one configuration of each model to compare their metric scores, consistency across drafts and per-script scores. Each menu starts on the model's best configuration.
Scores by Metric
GPT-5.6 Terra (ultra) scores higher on six of the nine metrics and Claude Opus 4.8 (max effort) on three. Weighted by each metric's share of the Writing Score, Substance, Accuracy & Value does the most to put GPT-5.6 Terra (ultra) ahead, with a 4.4-point lead on that metric. The weight beside each metric is its share of the Writing Score.
Consistency Across Drafts
| GPT-5.6 Terra (ultra) | Claude Opus 4.8 (max effort) | |
|---|---|---|
| Score spread across drafts (±) | 1.38 | 2.02 |
| Lowest script score | 79.8 Script 4 | 77.5 Script 4 |
| Highest script score | 83.0 Script 6 | 83.1 Script 6 |
| Range across scripts | 3.2 | 5.6 |
Spread is how much the Writing Score moves between a configuration's drafts, and lower is steadier. Green marks the better number in each row.
Scores by Script
Writing Score on each reference script. Gap is GPT-5.6 Terra (ultra) minus Claude Opus 4.8 (max effort), and the higher score is in bold.
| Script | GPT-5.6 Terra (ultra) | Claude Opus 4.8 (max effort) | Gap |
|---|---|---|---|
| Script 1 opinion / warning explainer | 81.1 | 80.2 | +0.9 |
| Script 2 news-analysis / skeptical explainer | 82.1 | 80.6 | +1.5 |
| Script 3 personal roadmap / opinion | 82.5 | 79.3 | +3.2 |
| Script 4 short explainer | 79.8 | 77.5 | +2.3 |
| Script 5 news-analysis / opinion explainer | 82.2 | 80.3 | +1.9 |
| Script 6 founder announcement / personal origin story | 83.0 | 83.1 | -0.1 |
| Script 7 personal technical walkthrough / agentic workflow case study | 80.6 | 82.5 | -1.9 |
| Script 8 product release announcement / personal observation | 80.3 | 82.5 | -2.2 |
| Script 9 career guide / hiring analysis | 81.8 | 82.0 | -0.2 |
| Script 10 engineering process walkthrough / presentation adaptation | 80.6 | 78.9 | +1.7 |
Scorecards and Thinking Pages
Each configuration name in the table above opens its own scorecard.
Compare With Another Model
Pick a second model. A pair with its own written match-up opens that page, and any other pair opens in the comparison tool.
Nearby Match-Ups by Best Writing Elo
Every configuration wrote five drafts for each of the same 10 briefs, from the same prompt. Generation was closed-book, with the prompt as the only source and no web search or other tools. Each judge scores every draft once: Claude Opus 5.5 at medium effort through the Claude Code subscription, GPT-6.1 Sol at medium reasoning through the Codex CLI, GLM 5.3 Flash through OpenRouter, and Jev 1.13, a typed judge from TypeSafe that answers each metric as a multiple-choice question. The channel's finished script is their reference for the voice, and it gets no score on the board itself.
← All comparisons