co Cohere Command R
Cohere · open weights · writing benchmark
The short version
Cohere Command R sits at #125 of 130 on ToneBench, toward the value end of the board, with a writing Elo of 263 and an overall score of 51.3 out of 100. We measured it by having it write all 9 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 Cohere it ranks 5th of 6. Cohere Command A Reasoning is the family's top writer here, about 353 Elo ahead, so this config trades some quality for whatever it saves you in cost or speed.
Among open-weights models, it comes in 42nd of 47. Open weights still trail the closed frontier on pure voice fidelity, and you can see that in the gap at the top.
Worth knowing before you rely on it: this model is swingy. Its overall score moves ± 9.1 points between runs of the same brief, versus a board median of ± 3.8. A great draft and a mediocre one can come from the identical prompt. The shaded bands on the metric bars below show where that volatility lives.
What it does best is Anti-Slop: it ranks 112th on that dimension at 69.5, well below the board average. Its softer spot is Tone & Voice (53.0, 124th), which is the thing to watch if that metric matters most for your use.
It was uneven across the 9 scripts. Its best run was the founder announcement / personal origin story (66.5) and its weakest was the personal technical walkthrough / agentic workflow case study (40.0), a spread of about 27 points. Worth knowing if your writing skews toward one of those formats.
On cost, it runs about $0.002/task. Cohere Command A Reasoning scores higher for less money, at $0/task, so on pure value this config is not the frontier.
Skill profile
The shape is the story: the further a corner reaches, the stronger that dimension. Biggest edge over the board: Anti-Slop (-13.2); furthest behind: Length Discipline (-41.4).
Per-metric scores
Nine writing dimensions, each scored 0–100 and blended by the editorial weight shown. Rank is against all 130 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 9 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, Cohere Command R varies by ± 9.1 points between runs versus a board median of ± 3.8, so it is swingier than the typical model here, worth knowing if you need repeatable output.
Its most volatile dimension is Tone & Voice (± 3.8 vs a board median of ± 2.2): two runs of the same brief can land visibly different tone & voice scores. Its steadiest is Length Discipline (± 4.4 vs ± 8.6 board median), which you can rely on run after run.
| Metric | This model | Board median | Verdict |
|---|---|---|---|
| Tone & Voice ? | 53.0 ± 3.8 | ± 2.2 | swingier than most |
| Writing Craft ? | 61.1 ± 3.0 | ± 2.0 | typical |
| Substance & Value ? | 60.1 ± 4.2 | ± 2.6 | swingier than most |
| Flow & Emotion ? | 48.6 ± 4.1 | ± 2.6 | swingier than most |
| YouTube Structure ? | 41.6 ± 5.5 | ± 3.6 | swingier than most |
| Hook ? | 52.6 ± 7.4 | ± 3.0 | swingier than most |
| Length Discipline ? | 25.8 ± 4.4 | ± 8.6 | steadier than most |
| Anti-Slop ? | 69.5 ± 5.3 | ± 2.6 | swingier than most |
| Visual Cues ? | 42.2 ± 10.4 | ± 4.7 | swingier than most |
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 51.3 is the consensus of three family-disjoint judges scoring the same 45 stored drafts blind with the identical rubric. The highest and lowest judge differ by 11.8 points on its overall. A typical amount of judge disagreement for this board. How the panel works: methodology.
How we ran Cohere Command R
- Accessed viaCohere API · model page ↗
- Exact model / route id
command-r-08-2024(provider default reasoning) - Generations5 independent runs × 9 scripts = 45 drafts, stored immutably after generation
- Run dates2026-07-29 → 2026-08-11
- Judges3-judge panel: Claude Opus 5, high effort (Anthropic) + DeepSeek V4 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 Cohere Command R at writing?
On ToneBench it ranks #125 of 130 with a writing Elo of 263 and an overall score of 51.3/100. That score comes from writing our 9 real YouTube scripts five times each and scoring every draft blind against our own finished versions across nine writing dimensions.
Is Cohere Command R the best Cohere model for writing?
Not quite. Within Cohere it ranks 5th of 6; Cohere Command A Reasoning is the family's best writer here.
Is Cohere Command R good value for the money?
It costs about $0.002/task. Cohere Command A Reasoning scores higher for less, so it is not the value pick.
What are Cohere Command R's strengths and weaknesses?
Its strongest dimension is Anti-Slop (112th on the board, 69.5). Its weakest is Tone & Voice (124th on the board, 53.0). The full nine-metric breakdown is on this page.
Does Cohere Command R write some formats better than others?
Yes. Its best of our 9 scripts was the founder announcement / personal origin story (66.5) and its weakest was the personal technical walkthrough / agentic workflow case study (40.0).
How consistent is Cohere Command R between runs?
We run every script 5 times. Cohere Command R's overall score varies by about ±9.1 points between runs, versus a board median of ±3.8. That is swingier than typical, so expect more draft-to-draft variation. The full per-metric variability table is on this page.
How was Cohere Command R evaluated?
Via Cohere API using the exact model/route id command-r-08-2024, run on 2026-07-29. 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 Cohere Command R open source?
Yes, it is an open-weights model. Among open-weights models it ranks 42nd of 47 for writing.