Towards AITowards AIToneBench

Model Comparison

GPT-6.1 Sol vs Claude Haiku 5.5

GPT-6.1 Sol (max) leads Claude Haiku 5.5 (max effort) by 239 Elo, and their 95% ranges sit 158 Elo apart. Among 87 models, each ranked by its best configuration, GPT-6.1 Sol is #8 and Claude Haiku 5.5 is #20. Cost per script: $0.239 for GPT-6.1 Sol (max) and $0.103 for Claude Haiku 5.5 (max effort).

Last updated , from board data generated at 17:52 UTC.

GPT-6.1 Sol
2153 Writing Elo of GPT-6.1 Sol (max), its best configuration#8 of 87 models · 82.4/100 Writing Score · $0.239 per script5 configurations on the board
Claude Haiku 5.5
1914 Writing Elo of Claude Haiku 5.5 (max effort), its best configuration#20 of 87 models · 80.8/100 Writing Score · $0.103 per script5 configurations on the board

Picks and Verdict

PickGPT-6.1 Sol

GPT-6.1 Sol (low)

Pick GPT-6.1 Sol (low) as the default, since at $0.052 a script in 1.5 minutes it rates above Haiku's max effort, with separate ranges, at about half the price.

PickClaude Haiku 5.5

Claude Haiku 5.5 (xhigh)

Pick Claude Haiku 5.5 (xhigh) for bulk drafts at $0.029 a script, because it already holds most of what Haiku gains from thinking and finishes in 3.6 minutes.

Claude Haiku 5.5 at its most expensive setting doesn't reach GPT-6.1 Sol's cheapest one. Claude Haiku 5.5 (max effort) costs $0.103 a script and takes more than twelve minutes. GPT-6.1 Sol (low) costs $0.052, finishes in 1.5 minutes and rates 95 Elo higher. Their 95% ranges sit just apart, so the board puts low ahead, if only just, and I see no reason to pay double and wait eight times as long for a draft that rates lower. Best against best, Sol leads by 239 Elo, the #8 model of 87 against the #20 model, each ranked by its best setting. Sol's medium and high rate below low inside overlapping ranges, so I'd skip their extra cost, while max adds 144 Elo over low, with separate ranges, at about 4.6 times the price. Haiku climbs much harder: it gains 482 Elo from low to max effort, and most of that arrives by xhigh. Claude Haiku 5.5 (xhigh) costs $0.029, and max effort adds 68 Elo for about 3.6 times the price, with ranges that overlap. Haiku's writing craft comes within a few tenths of Sol's at the top settings, and it loses on the hook. For me it's Sol at low, and Haiku at xhigh when every cent counts.

Disclosure: GPT-6.1 Sol and Claude Haiku 5.5 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-6.1 Sol and Claude Haiku 5.5, scored on the same 10 scripts by the same four judges.

How to Read the Chart
  • 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).
GPT-6.1 SolClaude Haiku 5.5
ConfigurationWriting Elo95% RangeWriting ScoreCost / ScriptTime / ScriptConfiguration Rank
GPT-6.1 Sol (max)21532118 to 219182.4$0.23910.2 min#18
GPT-6.1 Sol (xhigh)21362097 to 217982.4$0.1997.8 min#23
GPT-6.1 Sol (low)20091976 to 204381.4$0.0521.5 min#31
GPT-6.1 Sol (high)19751936 to 201381.1$0.0943.3 min#33
GPT-6.1 Sol (medium)19461915 to 198381.0$0.0541.6 min#37
Claude Haiku 5.5 (max effort)19141874 to 196080.8$0.10312.6 min#42
Claude Haiku 5.5 (xhigh)18461804 to 188780.0$0.0293.6 min#56
Claude Haiku 5.5 (high)16481603 to 169578.0$0.0121.5 min#71
Claude Haiku 5.5 (medium)15511505 to 159376.7$0.00670.8 min#84
Claude Haiku 5.5 (low)14321383 to 147575.2$0.00450.5 min#96

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.

Writing Elo at Each Shared Level Name

LevelGPT-6.1 SolClaude Haiku 5.5Gap95% Ranges
low20091432+577Separate
medium19461551+395Separate
high19751648+327Separate
xhigh21361846+290Separate
max21531914+239Separate

Gap is GPT-6.1 Sol minus Claude Haiku 5.5 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

GPT-6.1 Sol (max)Claude Haiku 5.5 (max effort)Solid bar: higher score in that row

GPT-6.1 Sol (max) scores higher on all nine metrics. Weighted by each metric's share of the Writing Score, Hook Strength does the most to put GPT-6.1 Sol (max) ahead, with a 4.6-point lead on that metric. The weight beside each metric is its share of the Writing Score.

Tone & Voice Match19% weight
78.1
76.4
Writing Craft & Clarity13% weight
78.1
77.8
Substance, Accuracy & Value15% weight
87.3
85.6
Continuity & Emotion14% weight
79.2
78.7
YouTube Best Practices12% weight
78.6
76.8
Hook Strength10% weight
82.8
78.2
Length Adherence8% weight
98.0
97.3
Slop Score5% weight
88.2
86.0
Visual Cue Quality4% weight
83.0
80.9

Consistency Across Drafts

GPT-6.1 Sol (max)Claude Haiku 5.5 (max effort)
Score spread across drafts (±)1.401.55
Lowest script score80.6 Script 1078.2 Script 10
Highest script score85.0 Script 982.6 Script 4
Range across scripts4.44.4

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-6.1 Sol (max) minus Claude Haiku 5.5 (max effort), and the higher score is in bold.

ScriptGPT-6.1 Sol (max)Claude Haiku 5.5 (max effort)Gap
Script 1 opinion / warning explainer81.780.3+1.4
Script 2 news-analysis / skeptical explainer82.580.7+1.8
Script 3 personal roadmap / opinion83.680.8+2.8
Script 4 short explainer82.582.6-0.1
Script 5 news-analysis / opinion explainer82.081.7+0.3
Script 6 founder announcement / personal origin story83.381.1+2.2
Script 7 personal technical walkthrough / agentic workflow case study82.280.0+2.2
Script 8 product release announcement / personal observation81.081.3-0.3
Script 9 career guide / hiring analysis85.081.5+3.5
Script 10 engineering process walkthrough / presentation adaptation80.678.2+2.4

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