Model Comparison
Claude Haiku 5.5 vs GPT-6.1 Terra
Overlapping ranges: the best configurations, Claude Haiku 5.5 (max effort) and GPT-6.1 Terra (max), are 43 Elo apart, and their 95% ranges overlap. Among 84 models, each ranked by its best configuration, Claude Haiku 5.5 is #19 and GPT-6.1 Terra is #23. Cost per script: $0.103 for Claude Haiku 5.5 (max effort) and $0.079 for GPT-6.1 Terra (max).
Last updated , from board data generated at 20:15 UTC.
Picks and Verdict
PickClaude Haiku 5.5
Claude Haiku 5.5 (max effort)
Pick Claude Haiku 5.5 (max effort) when the script has to land on its target length, at $0.103 a script and 12.6 minutes of waiting.
PickGPT-6.1 Terra
GPT-6.1 Terra (low)
Pick GPT-6.1 Terra (low) for everyday drafts, since its individual Elo range overlaps Haiku's best and it costs $0.027 and about half a minute a script.
These two have overlapping individual Elo ranges at the top, and I'd choose on speed and length. Claude Haiku 5.5 is the #19 model of 84 and GPT-6.1 Terra the #23 model, each ranked by its best setting. Claude Haiku 5.5 (max effort) sits 43 Elo above GPT-6.1 Terra (max) with overlapping 95% ranges, and it overlaps Terra at low too. Their ladders couldn't be more different. Haiku climbs 489 Elo from low to max effort, and at medium Terra leads it by 307 Elo with separate ranges. Terra's five levels sit within 25 Elo of each other with overlapping ranges, so low is the one I would pick, at $0.027 and about half a minute a script. Haiku at max costs $0.103 and takes 12.6 minutes. Haiku at xhigh overlaps Terra at low for $0.029, but it still takes 3.6 minutes. The scores split cleanly. Haiku at max lands much closer to the target length, 9.3 points up on Length Adherence against Terra at low, with better visual cues, while Terra writes the stronger hook with less slop. For quick drafts I'd use Terra at low. When the length has to be right without trimming, Haiku at max is worth the wait.
Disclosure: Claude Haiku 5.5 and GPT-6.1 Terra each come from the same lab as one of the judges (Claude Opus 5.5 and GPT-6.1 Sol). 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 Claude Haiku 5.5 and GPT-6.1 Terra, 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.
- Each bar describes one configuration. Overlap alone does not establish a tie or decide their order. Comparing these ranges does not directly test the difference between a pair.
- 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 |
|---|---|---|---|---|---|---|
| Claude Haiku 5.5 (max effort) | 1966 | 1925 to 2006 | 80.8 | $0.103 | 12.6 min | #35 |
| GPT-6.1 Terra (max) | 1923 | 1884 to 1959 | 80.3 | $0.079 | 1.8 min | #43 |
| GPT-6.1 Terra (high) | 1922 | 1881 to 1958 | 80.3 | $0.027 | 0.5 min | #44 |
| GPT-6.1 Terra (low) | 1920 | 1879 to 1961 | 80.2 | $0.027 | 0.5 min | #46 |
| GPT-6.1 Terra (xhigh) | 1906 | 1869 to 1944 | 80.1 | $0.028 | 0.5 min | #48 |
| GPT-6.1 Terra (medium) | 1898 | 1852 to 1942 | 79.9 | $0.027 | 0.5 min | #49 |
| Claude Haiku 5.5 (xhigh) | 1895 | 1854 to 1937 | 80.0 | $0.029 | 3.6 min | #50 |
| Claude Haiku 5.5 (high) | 1686 | 1647 to 1733 | 78.0 | $0.012 | 1.5 min | #62 |
| Claude Haiku 5.5 (medium) | 1591 | 1537 to 1636 | 76.7 | $0.0067 | 0.8 min | #74 |
| Claude Haiku 5.5 (low) | 1477 | 1418 to 1522 | 75.2 | $0.0045 | 0.5 min | #86 |
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. The 95% ranges describe individual configurations; overlap alone establishes neither a tie nor an order for a pair.
- Best Configurations Head to Head. Claude Haiku 5.5 (max effort) at Elo 1966 (range 1925 to 2006) and GPT-6.1 Terra (max) at Elo 1923 (range 1884 to 1959) have overlapping 95% ranges.
- Lowest Thinking Levels Head to Head. At their lowest thinking levels, GPT-6.1 Terra (low) at Elo 1920 and $0.027 per script leads Claude Haiku 5.5 (low) at Elo 1477 and $0.0045 per script by 443 Elo. Their 95% ranges do not overlap.
- Same Level Name, Side by Side. Across the level names both models use, Claude Haiku 5.5 leads at max; GPT-6.1 Terra leads at low, medium, high, and xhigh. The 95% ranges do not overlap at low, medium, and high. Level names come from each provider, so the same name does not promise the same reasoning budget.
- Cheapest Configuration Overlapping the Top One. GPT-6.1 Terra (high), at $0.027 per script, is the cheapest configuration whose 95% range overlaps that of Claude Haiku 5.5 (max effort), at $0.103 per script: 1881 to 1958 against 1925 to 2006. Claude Haiku 5.5 (max effort) costs about 3.8 times as much per script.
- Cheapest Claude Haiku 5.5 Configuration Above GPT-6.1 Terra's Best. The cheapest Claude Haiku 5.5 configuration with a higher Writing Elo than GPT-6.1 Terra (max), at Elo 1923 and $0.079 per script, is Claude Haiku 5.5 (max effort), at Elo 1966 and $0.103 per script. Their 95% ranges overlap. It costs about 1.3 times as much per script.
- Lowest to Highest Thinking Level. Claude Haiku 5.5 moves +489 Elo from low to max effort, at about 23 times the cost per script. GPT-6.1 Terra moves +3 Elo from low to max, at about 2.9 times the cost per script.
Writing Elo at Each Shared Level Name
| Level | Claude Haiku 5.5 | GPT-6.1 Terra | Gap | 95% Ranges |
|---|---|---|---|---|
| low | 1477 | 1920 | -443 | Separate |
| medium | 1591 | 1898 | -307 | Separate |
| high | 1686 | 1922 | -236 | Separate |
| xhigh | 1895 | 1906 | -11 | Overlap |
| max | 1966 | 1923 | +43 | Overlap |
Gap is Claude Haiku 5.5 minus GPT-6.1 Terra at the same level name, using each model's highest-rated configuration at that level.
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
Claude Haiku 5.5 (max effort) scores higher on five of the nine metrics and GPT-6.1 Terra (max) on four. Weighted by each metric's share of the Writing Score, Tone & Voice Match does the most to put Claude Haiku 5.5 (max effort) ahead, with a 2.4-point lead on that metric. The weight beside each metric is its share of the Writing Score.
Consistency Across Drafts
| Claude Haiku 5.5 (max effort) | GPT-6.1 Terra (max) | |
|---|---|---|
| Score spread across drafts (±) | 1.55 | 1.68 |
| Lowest script score | 78.2 Script 10 | 77.9 Script 10 |
| Highest script score | 82.6 Script 4 | 83.2 Script 9 |
| Range across scripts | 4.4 | 5.3 |
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 Claude Haiku 5.5 (max effort) minus GPT-6.1 Terra (max), and the higher score is in bold.
| Script | Claude Haiku 5.5 (max effort) | GPT-6.1 Terra (max) | Gap |
|---|---|---|---|
| Script 1 opinion / warning explainer | 80.3 | 80.2 | +0.1 |
| Script 2 news-analysis / skeptical explainer | 80.7 | 80.7 | 0.0 |
| Script 3 personal roadmap / opinion | 80.8 | 80.4 | +0.4 |
| Script 4 short explainer | 82.6 | 79.6 | +3.0 |
| Script 5 news-analysis / opinion explainer | 81.7 | 80.1 | +1.6 |
| Script 6 founder announcement / personal origin story | 81.1 | 82.2 | -1.1 |
| Script 7 personal technical walkthrough / agentic workflow case study | 80.0 | 79.9 | +0.1 |
| Script 8 product release announcement / personal observation | 81.3 | 79.1 | +2.2 |
| Script 9 career guide / hiring analysis | 81.5 | 83.2 | -1.7 |
| Script 10 engineering process walkthrough / presentation adaptation | 78.2 | 77.9 | +0.3 |
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