Claude Sonnet 5.5 ranks #3 of 87 models on the November 2026 ToneBench board through its best configuration, Claude Sonnet 5.5 (max effort). This setting, Claude Sonnet 5.5 (xhigh), is #15 of 151 configurations, with a Writing Score of 83.2/100 and a Writing Elo of 2262. One script costs $0.302 at API list price.
Last updated , from board data generated at 17:52 UTC.
Rank, scores, the nine metrics, the 10 reference scripts, cost and setup for one
configuration (a model at one reasoning setting). Key Results sums it up; the section links below it jump to the detail.
Model Rank ?
#3 of 87 modelsthrough Claude Sonnet 5.5 (max effort); this setting is #15 of 151 configurations
Writing Score ?
83.2 out of 100
Writing Elo ?
2262 95% range 2215 to 2299
Cost per Script
$0.302 at API list price
Key Results
StandingClaude Sonnet 5.5 ranks #3 of 87 models through its best configuration, Claude Sonnet 5.5 (max effort). This setting is #15 of 151 configurations by Writing Elo (2262), in the top 20. Its 95% range (2215 to 2299) overlaps 4 other configurations, ranked #13 to #17.
MetricsBest-ranked: Writing Craft & Clarity (80.8, 12th of 151 configurations, +5.8 against the board median). Lowest-ranked: Hook Strength (81.2, 46th of 151 configurations, +1.5 against the board median).
ScriptsFrom 81.4 on Script 10 (engineering process walkthrough) to 84.5 on Script 9 (career guide): a 3.1-point spread across the 10 reference scripts.
ConsistencyIts Writing Score moves ± 1.2 between drafts of the same script, against ± 2.5 for the median configuration: steadier than most.
ValueGPT-6 Sol (max) ranks higher with an overlapping Elo range and costs about $0.126 per script, against $0.302 per script for this one.
Writing Elo comes from comparing its average on each script with every other configuration's average on that same script; a gap smaller than the draft-to-draft variation counts as a draw (what the ranges can and cannot show).
On Writing Craft & Clarity, its best-ranked metric, its sentences need less smoothing than the typical configuration's. Even its lowest-ranked metric, Hook Strength, ranks in the top half of the board, so no metric stands out as a weak spot.
The script format moves its score. If most of your scripts look like its best or weakest format, that result is a better guide than the average.
Its drafts vary little, so the first draft is usually close to the next one.
It outranks 10 configurations that cost at least 50% more.
Among proprietary models, Claude Sonnet 5.5 is 3rd of 53. This setting is 15th of 108 proprietary configurations. It also sits above every open-weight configuration; the best of them, GLM-5.3 (max), is configuration #22.
Lab
Anthropic 3rd of 13 modelsthis setting 14th of 34 configurations
Weights
Proprietary 3rd of 53 modelsthis setting 15th of 108 configurations
Consistency
± 1.2 steadier than most
Average Tokens
40.8k input + output
Time per Script
3.1 min average, retries included
Highest-Ranked in This LabClaude Opus 5.5 (xhigh) ranks highest of the Anthropic configurations by Writing Elo, with a Writing Score +2.6 against this configuration, for about 70 cents more per script.
Each corner is one metric on a 0 to 100 scale; the further out, the higher the score. A corner outside the dashed line beats the board median. Hover or tap a point for its exact score, and a metric name for what it rewards.
Claude Sonnet 5.5 (xhigh)Board medianBoard best per metricMax possible (100)
Scores by Metric
Nine metrics, each scored 0 to 100 and weighted into the Writing Score; the weight sits under each name. Rank is out of all 151 ranked configurations.
Score colors:Strong: 82 and upMiddle: 65 to 81.9Weak: under 65
This modelBoard best (upper line)Board median (lower line)±: spread between drafts?: what the metric rewards
Tone & Voice Match ?19% weight · +5.5 vs median
board best 82.8 · Claude Opus 5.5 (max effort)79.5± 1.6 · 22ndboard median 74.0
Writing Craft & Clarity ?13% weight · +5.8 vs median
board best 83.2 · Claude Opus 5.5 (xhigh)80.8± 0.9 · 12thboard median 75.0
Substance, Accuracy & Value ?15% weight · +4.8 vs median
board best 88.4 · Claude Opus 5.5 (xhigh)86.6± 1.0 · 21stboard median 81.8
Continuity & Emotion ?14% weight · +5.2 vs median
board best 85.7 · Claude Opus 5.5 (max effort)82.7± 1.0 · 18thboard median 77.5
YouTube Best Practices ?12% weight · +5.5 vs median
board best 84.5 · Claude Fable 5.2 (max effort)81.8± 1.8 · 18thboard median 76.4
Hook Strength ?10% weight · +1.5 vs median
board best 84.0 · Claude Fable 5.2 (xhigh)81.2± 2.1 · 46thboard median 79.6
Length Adherence ?8% weight · +9.7 vs median
board best 99.0 · Claude Fable 5.2 (xhigh)94.5± 3.3 · 26thboard median 84.8
Slop Score ?5% weight · +5.3 vs median
board best 89.0 · GPT-6 Sol (medium)86.5± 1.0 · 27thboard median 81.3
Visual Cue Quality ?4% weight · +5.4 vs median
board best 85.9 · Claude Opus 5.5 (max effort)80.2± 2.8 · 33rdboard median 74.8
Scores by Reference Script
Every ranked configuration writes five drafts for each of the same 10 briefs, and each score is the mean of those drafts. The rank and board median compare it with every configuration on that script, which separates a weak result from a hard script. The result on the format closest to your own work is the most relevant.
Script 1
opinion / warning explainer
84.0
out of 100
5th of 151 configurations on this script · board median 77.2
Script 2
news-analysis / skeptical explainer
82.8
out of 100
21st of 151 configurations on this script · board median 78.4
Script 3
personal roadmap / opinion
82.9
out of 100
20th of 151 configurations on this script · board median 77.2
Script 4
short explainer
82.7
out of 100
15th of 151 configurations on this script · board median 75.7
Script 5
news-analysis / opinion explainer
83.9
out of 100
14th of 151 configurations on this script · board median 78.0
Script 6
founder announcement / personal origin story
84.0
out of 100
15th of 151 configurations on this script · board median 79.2
Script 7
personal technical walkthrough / agentic workflow case study
83.4
out of 100
21st of 151 configurations on this script · board median 77.9
Script 8
product release announcement / personal observation
82.7
out of 100
18th of 151 configurations on this script · board median 79.0
Script 9
career guide / hiring analysis
84.5
out of 100
20th of 151 configurations on this script · board median 79.3
Script 10
engineering process walkthrough / presentation adaptation
81.4
out of 100
18th of 151 configurations on this script · board median 75.0
Consistency Between Drafts
Every score on this page is the mean of 5 drafts per script. The ± is the standard deviation between those drafts: the larger it is, the more two drafts of the same script can differ.
Compared with the typical model, it's steadiest on Substance, Accuracy & Value: ± 1.0 between drafts against ± 2.1.
Consistency by Metric
Metric
This model
Board median
Verdict
Tone & Voice Match ?
79.5 ± 1.6
± 1.9
typical
Writing Craft & Clarity ?
80.8 ± 0.9
± 1.5
typical
Substance, Accuracy & Value ?
86.6 ± 1.0
± 2.1
steadier than most
Continuity & Emotion ?
82.7 ± 1.0
± 1.5
typical
YouTube Best Practices ?
81.8 ± 1.8
± 2.4
typical
Hook Strength ?
81.2 ± 2.1
± 2.7
typical
Length Adherence ?
94.5 ± 3.3
± 6.6
steadier than most
Slop Score ?
86.5 ± 1.0
± 1.6
typical
Visual Cue Quality ?
80.2 ± 2.8
± 2.9
typical
Cost, Time and Tokens per Script
Logged on every run; none of these figures affects the Writing Score. Time is the average time to one accepted script, retries included. Cost is one script at the API list price.
Time per Script
3.1 minabout 2x the board median of 1.5 min
Prompt tokens in
13.2kstyle guide + brief + research packet
Tokens Out
27.6kwell above the board median, so a lot of thinking
Cost per Script
$0.302provider-counted output, thinking included, plus the fixed prompt priced once; varies ± 0.167 between drafts
List Price Used
$2 / $10input / output, per million tokens
Scores by Judge
Four judges score the same 50 stored drafts blind against the same rubric: three LLM judges from three labs, plus Jev, a typed judge from TypeSafe that rates each metric on five described quality levels. No single judge's taste decides the result. The Writing Score of 83.2 combines their scores with the parts computed in code (Length Adherence and half of the Slop Score). The highest and lowest judge are 9.6 points apart on its Writing Score. That is tighter than the board median of 13.4, so the judges mostly agree on this one. Each judge also has its own scale, so every card shows that judge's score against its average across the board. How the panel works.
Claude Opus 5.5 (medium)
Anthropic
81.4
+12.7 against its board average of 68.7
GLM 5.3 Flash
Z.ai
90.1
+7.3 against its board average of 82.9
GPT-6.1 Sol (medium)
OpenAI
80.5
+8.6 against its board average of 71.9
Jev 1.13
TypeSafe
80.9
+6.8 against its board average of 74.0
Evaluation Setup
How this exact configuration was run and judged: route, model ID, dates, sampling and prompt.
Accessed viaClaude Code (Anthropic subscription).
Exact model / route idclaude-sonnet-5-5 at reasoning level: xhigh.
Generations5 independent runs × 10 scripts = 50 drafts, stored after generation and never edited.
Run dates2026-10-08.
JudgesClaude 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 from TypeSafe. One call per judge per draft. All four score every draft on their own, blind to which model wrote it. The three LLM judges use the identical rubric and have to quote evidence before they score a metric. Jev answers the same rubric as one question per metric, choosing among five described quality levels, and writes no text. Seven metrics are the plain average of the four judges. Length is computed from the spoken word count, and Slop is half computed (lexical scans minus the spoken-dash penalty) and half judged. The Writing Elo is cross-checked with a Bradley-Terry fit of the same pairwise results (methodology).
Sampling and promptThe provider's default sampling, with no fine-tuning and no system-prompt changes. Every current-ranked model gets the same prompt, about 13,000 tokens (the count shifts a bit with each model's tokenizer): a style guide built from Louis's voice and storytelling guides, the target length, and the brief with its research packet. Generation is closed-book on every route: the prompt is the only source, with no web search, browsing, code execution, file access or other tool (generation).
What stays privateThe exact rubric and any non-public reference material. Each script is normally benchmarked before its video is published, which lowers the odds that a model has seen the reference, though it does not rule it out (why).
Frequently Asked Questions
How Good Is Claude Sonnet 5.5 (xhigh) at Writing?
Claude Sonnet 5.5 ranks #3 of 87 models on ToneBench through its best configuration, Claude Sonnet 5.5 (max effort). This configuration is #15 of 151 configurations, in the top 20, with a Writing Elo of 2262 and a Writing Score of 83.2/100. The ranks come from full YouTube scripts written in the voice of Louis-François Bouchard and judged blind, with the channel's finished script as the reference. The benchmark measures long-form scripted writing, so the result does not carry over to every kind of text.
Is Claude Sonnet 5.5 (xhigh) the Best Anthropic Model for Writing?
No. Claude Sonnet 5.5 is 3rd of 13 Anthropic models. The top one is Claude Opus 5.5, through Claude Opus 5.5 (xhigh), about 442 Elo ahead of this setting.
Is Claude Sonnet 5.5 (xhigh) Good Value for the Money?
It costs about $0.302 per script. GPT-6 Sol (max) ranks higher with an overlapping Elo range and costs less.
What Are the Strengths and Weaknesses of Claude Sonnet 5.5 (xhigh)?
Its best-ranked metric is Writing Craft & Clarity (80.8, 12th of 151 configurations, +5.8 against the board median), and its lowest-ranked is Hook Strength (81.2, 46th of 151, +1.5). Even that one ranks in the top half of the board. The Scores by Metric section shows the board median and the board's best for each one.
How Was Claude Sonnet 5.5 (xhigh) Evaluated?
It was run through Claude Code (Anthropic subscription), with the exact model/route ID claude-sonnet-5-5 at reasoning effort xhigh, first run on 2026-10-08. It wrote each script 5 times with the provider's default sampling and no fine-tuning. The Evaluation Setup section and the methodology page have the details.
Does Claude Sonnet 5.5 (xhigh) Have Open Weights?
No. It is a proprietary model: its weights are not public. Among proprietary configurations it is 15th of 108 for writing.