Coding & Engineering: Weakest Dimension (2026)
Each Coding & Engineering tool's lowest recorded dimension among the five rubric axes, plus how often each dimension lands at the bottom across the channel (55 tools). Ties are resolved to the earlier rubric axis (quality → features → usability → performance → value) and disclosed here.
Weakest-Dimension Census
Number of tools whose minimum recorded dimension falls on each axis.
Per-Tool Weakest Dimension (Recorded)
| Tool | Weakest dimension | Value | Overall |
|---|---|---|---|
| Cursor | Value for Money | 82 | 89.2 |
| GitHub Copilot | Performance | 85 | 88.1 |
| Claude Code | Usability | 75 | 86.1 |
| Zed | Feature Depth | 84 | 86.1 |
| Aider | Usability | 72 | 86.0 |
| OpenAI Codex | Value for Money | 74 | 84.6 |
| Amazon Q Developer | Value for Money | 80 | 84.3 |
| Trae | Output Quality | 80 | 84.2 |
| Kilo | Performance | 82 | 84.0 |
| Bito AI | Usability | 82 | 83.9 |
| Windsurf | Value for Money | 78 | 83.9 |
| Cline | Usability | 76 | 83.7 |
| JetBrains AI Assistant | Value for Money | 78 | 83.4 |
| Sourcery | Feature Depth | 82 | 83.4 |
| CodeRabbit | Value for Money | 76 | 83.2 |
| Korbit AI | Feature Depth | 81 | 82.8 |
| Tabby | Feature Depth | 76 | 82.6 |
| Tabnine | Value for Money | 78 | 82.5 |
| Continue | Feature Depth | 78 | 82.3 |
| Greptile | Value for Money | 76 | 82.2 |
| Amp | Value for Money | 78 | 82.1 |
| Codestral | Usability | 78 | 82.0 |
| Warp | Value for Money | 74 | 81.9 |
| Replit Agent | Performance | 78 | 81.8 |
| Zencoder | Value for Money | 74 | 81.7 |
| Graphite | Value for Money | 76 | 81.6 |
| OpenHands | Usability | 76 | 81.5 |
| Supermaven | Feature Depth | 74 | 81.4 |
| Refact.ai | Feature Depth | 78 | 81.2 |
| Augment Code | Value for Money | 70 | 80.7 |
| Qodo | Value for Money | 74 | 80.6 |
| Snyk Code | Value for Money | 74 | 80.6 |
| Moderne | Value for Money | 70 | 80.4 |
| Poolside | Usability | 76 | 80.4 |
| Jules | Feature Depth | 78 | 79.7 |
| Sourcegraph Cody | Value for Money | 72 | 79.7 |
| Sonar AI Code Assurance | Value for Money | 72 | 79.5 |
| Qodana | Value for Money | 76 | 79.4 |
| Roomote | Value for Money | 76 | 79.3 |
| Devin | Value for Money | 68 | 78.5 |
| Sweep | Feature Depth | 76 | 77.8 |
| CodeBuddy | Feature Depth | 76 | 77.6 |
| Ellipsis | Value for Money | 74 | 77.6 |
| Grit | Usability | 70 | 77.6 |
| Tongyi Lingma | Feature Depth | 76 | 77.6 |
| Pieces for Developers | Value for Money | 70 | 77.6 |
| CodeGeeX | Feature Depth | 74 | 77.1 |
| Baidu Comate | Feature Depth | 74 | 76.7 |
| Codeflash | Feature Depth | 74 | 76.4 |
| Mutable AI | Value for Money | 74 | 76.1 |
| Double | Value for Money | 72 | 75.1 |
| Charlie | Value for Money | 66 | 73.2 |
| Blackbox AI | Value for Money | 70 | 72.4 |
| Magic | Value for Money | 60 | 71.6 |
| Codegen | Value for Money | 62 | 70.9 |
Methodology — all figures are computed on this page from the published archives only (data/tools, 55 recorded coding tools, snapshots 2026-09-06 – 2026-09-17). Medians use the standard median (middle element, or the mean of the two middle elements rounded to one decimal); means are arithmetic averages rounded to one decimal. A low dimension marks a recorded rubric minimum, not a product defect. No figure on this page is sampled, estimated or sponsored.