Claude Opus 5.5 (low)
Anthropic · proprietary · writing benchmark
The short version
Claude Opus 5.5 (low) sits at #10 of 167 on ToneBench, in the upper tier, with a writing Elo of 2320 and an overall score of 89.1 out of 100. We measured it by having it write all 10 of our real YouTube scripts, five times each, then scoring every draft blind against our own finished versions. Here is how it shook out.
Within Anthropic it ranks 10th of 29. Claude Opus 5.5 (max effort) is the family's top writer here, about 526 Elo ahead, so this config trades some quality for whatever it saves you in cost or speed.
Among proprietary models, it comes in 10th of 114. That ranking is against the stronger half of the board: closed models still set the pace on voice fidelity here.
What it does best is Substance & Value: it ranks 10th on that dimension at 88.8, well above the board average. Its softer spot is Anti-Slop (91.9, 28th), which is the thing to watch if that metric matters most for your use.
It was uneven across the 10 scripts. Its best run was the personal technical walkthrough / agentic workflow case study (90.8) and its weakest was the engineering process walkthrough / presentation adaptation (85.7), a spread of about 5 points. Worth knowing if your writing skews toward one of those formats.
On cost, it runs about $0.17/task. For that money it beats 14 models that cost noticeably more.
If you're tuning reasoning effort, we also tested Claude Opus 5.5 at other settings. The strongest of those on the board is Claude Opus 5.5 (max effort) (Elo 2846), so it is worth checking whether more or less thinking moves the needle before you lock in this one.
Skill profile
The shape is the story: the further a corner reaches, the stronger that dimension. Biggest edge over the board: Length Discipline (+19.1); smallest edge: Hook (+7.7).
Per-metric scores
Nine writing dimensions, each scored 0–100 and blended by the editorial weight shown. Rank is against all 167 current-ranked models. The thick bar is this model; the thin lines above and below are the current field's best model (darker) and average (lighter) on the same scale. The small ± number is its run-to-run variation.
Per-article scores
The same 10 real scripts every current-ranked model writes, scored individually. Different formats stress different skills.
Consistency: run-to-run variability
Every cell on this page is the mean of 5 independent runs per script; the ± numbers are the run-to-run standard deviation. Overall, Claude Opus 5.5 (low) varies by ± 1.6 points between runs versus a board median of ± 2.7, so it is noticeably steadier than the typical model here.
Its steadiest is Writing Craft (± 0.7 vs ± 1.5 board median), which you can rely on run after run.
| Metric | This model | Board median | Verdict |
|---|---|---|---|
| Tone & Voice ? | 88.7 ± 1.2 | ± 1.7 | typical |
| Writing Craft ? | 89.5 ± 0.7 | ± 1.5 | steadier than most |
| Substance & Value ? | 88.8 ± 0.9 | ± 2.0 | steadier than most |
| Flow & Emotion ? | 88.7 ± 0.7 | ± 1.8 | steadier than most |
| YouTube Structure ? | 89.1 ± 1.2 | ± 2.9 | steadier than most |
| Hook ? | 89.2 ± 1.2 | ± 2.0 | steadier than most |
| Length Discipline ? | 89.9 ± 4.9 | ± 8.6 | steadier than most |
| Anti-Slop ? | 91.9 ± 1.0 | ± 2.1 | steadier than most |
| Visual Cues ? | 87.7 ± 2.0 | ± 3.2 | typical |
Measured, not modeled
Numbers we log on every run and rarely talk about. None of these affect the writing scores; cost and latency are informational.
What each judge scored it
The published overall of 89.1 is the consensus of three family-disjoint judges scoring the same 50 stored drafts blind with the identical rubric. The highest and lowest judge differ by 4.2 points on its overall. The three judges essentially agree on this model. How the panel works: methodology.
How we ran Claude Opus 5.5 (low)
- Accessed viaClaude Code (Anthropic subscription)
- Exact model / route id
claude-opus-5-5atreasoning_effort: low - Generations5 independent runs × 10 scripts = 50 drafts, stored immutably after generation
- Run dates2026-09-25
- Judges3-judge panel: Claude Opus 5, high effort (Anthropic) + DeepSeek V4.1 Flash (OpenRouter) + GPT-5.6 Sol, medium reasoning (Codex CLI): every draft scored independently by all three with the identical rubric, blind to model identity, evidence quotes required before each metric score; published score = per-metric consensus mean, verified with a Bradley-Terry refit (methodology)
- Samplingprovider-default sampling, no fine-tuning, no system-prompt changes; every current-ranked model gets the identical brief, style contract and research packet
- What stays privatethe exact rubric and any non-public reference material; source-publication timing is disclosed for exceptions (why)
Frequently asked questions
How good is Claude Opus 5.5 (low) at writing?
On ToneBench it ranks #10 of 167 with a writing Elo of 2320 and an overall score of 89.1/100. That score comes from writing our 10 real YouTube scripts five times each and scoring every draft blind against our own finished versions across nine writing dimensions.
Is Claude Opus 5.5 (low) the best Anthropic model for writing?
Not quite. Within Anthropic it ranks 10th of 29; Claude Opus 5.5 (max effort) is the family's best writer here.
Is Claude Opus 5.5 (low) good value for the money?
It costs about $0.17/task. Nothing meaningfully cheaper outscores it, which puts it on the value side of the board.
What are Claude Opus 5.5 (low)'s strengths and weaknesses?
Its strongest dimension is Substance & Value (10th on the board, 88.8). Its weakest is Anti-Slop (28th on the board, 91.9). The full nine-metric breakdown is on this page.
Does Claude Opus 5.5 (low) write some formats better than others?
Yes. Its best of our 10 scripts was the personal technical walkthrough / agentic workflow case study (90.8) and its weakest was the engineering process walkthrough / presentation adaptation (85.7).
How consistent is Claude Opus 5.5 (low) between runs?
We run every script 5 times. Claude Opus 5.5 (low)'s overall score varies by about ±1.6 points between runs, versus a board median of ±2.7. That makes it one of the steadier models we test. The full per-metric variability table is on this page.
How was Claude Opus 5.5 (low) evaluated?
Via Claude Code (Anthropic subscription) using the exact model/route id claude-opus-5-5 at reasoning_effort low, run on 2026-09-25. 5 runs per script, provider-default sampling, no fine-tuning; every draft scored blind with a fixed rubric by a three-family judge panel. Full details in the 'How we ran it' section and the methodology.
Is Claude Opus 5.5 (low) open source?
No, it is a proprietary (closed-weights) model. Among proprietary models it ranks 10th of 114 for writing.