Coding & Engineering: Dimension Leaders (2026) — top tools per rubric dimension
Who leads each of the five rubric dimensions across all 55 recorded Coding & Engineering tools — the top three per dimension, each dimension's median and range, and where single-dimension leaders differ from the composite leader. Every figure is quoted from the published archives (channel overall median 81.4/100, as of 2026-09-17); no new scores are computed on this page.
Top 3 per Dimension (Recorded Scores)
Output Quality
median 84 · range 93–94Feature Depth
median 82 · range 92–92Usability
median 80 · range 90–90Performance
median 80 · range 88–88Dimension Leaders vs the Composite Leader
| Dimension | #1 by dimension score | Also composite top-3? |
|---|---|---|
| Output Quality | Claude Code (94) | Yes |
| Feature Depth | Cursor (92) | Yes |
| Usability | Cursor (90) | Yes |
| Performance | Cursor (88) | Yes |
| Value for Money | Aider (98) | No — leads this dimension only |
The composite leader of this channel is Cursor (89.2/100), followed by GitHub Copilot and Claude Code — ordered by the published overall score. 1 of 5 dimensions (Value for Money) is currently led by tool outside that composite top-3: leading a single dimension does not imply leading the composite, because the overall score weighs all five dimensions together (quality 0.25 + features 0.20 + usability 0.15 + performance 0.20 + value 0.20).
Methodology — all figures read only published archives (data/tools, dimsVersion v2-5dim, snapshots 2026-09-06 – 2026-09-17; 55 coding tools). The composite is the published deterministic weighting (quality 0.25 + features 0.20 + usability 0.15 + performance 0.20 + value 0.20); this page never recomputes or re-weights recorded scores, and medians use the standard median (middle element, or mean of the two middle elements rounded to 1 decimal). Entry prices are compared only within the same billing unit. No placement on this page is sponsored or affiliate-ranked; ordering is by recorded archive scores only.