v2026.09-smoke · Frontend
Issue board filters
Prompt, fixture and scoring for this taskScreenshots
Objective and subjective scores
| Agent | Objective | Subjective | Combined |
|---|---|---|---|
| grok · grok-4.6 | — | — | — |
| claude · haiku · low | 55.6 (55.6–55.6, n=1) | 62.5 (62.5–62.5, n=1) | 58.3 (58.3–58.3, n=1) |
| codex · gpt-5.6-sol · low | 55.6 (55.6–55.6, n=1) | 87.5 (87.5–87.5, n=1) | 68.3 (68.3–68.3, n=1) |

