AA Omniscience - Software Engineering (SWE) - HTML: leaderboard
HTML slice of AA Omniscience SWE, testing factual knowledge of markup, browser behavior, accessibility, and web engineering practice.
Metric: Accuracy (%). Source: artificialanalysis.ai. Saturation forecast: Around January 2027. 294 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi K3 (Max) | 82 | #30 (Kimi K3) |
| 2 | Kimi K3 (Low) | 82 | #30 (Kimi K3) |
| 3 | Claude Opus 4.5 (Thinking) | 78 | #83 (Claude Opus 4.5) |
| 4 | Gemini 3 Pro (Preview) (High) | 76 | #66 (Gemini 3 Pro (Preview)) |
| 5 | Claude Sonnet 4.5 (Thinking) | 72 | #142 (Claude Sonnet 4.5) |
| 6 | Claude Opus 4.5 (Non-reasoning) | 72 | #83 (Claude Opus 4.5) |
| 7 | Kimi K2.7 Code | 68 | #89 |
| 8 | Inkling (xHigh) | 68 | #126 (Inkling) |
| 9 | Kimi K2.6 | 66 | #104 |
| 10 | Kimi K2.5 (Thinking) | 66 | #144 (Kimi K2.5) |
| 11 | Hy3 | 64 | #93 |
| 12 | Motif 3 | 64 | #190 |
| 13 | Gemini 2.5 Pro | 58 | #152 |
| 14 | GLM-5.1 (Non-reasoning) | 58 | #97 (GLM-5.1) |
| 15 | GPT-5.1 (High) | 56 | #135 (GPT-5.1) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=aa-omniscience-software-engineering-swe-html · How It Works · Data refreshed daily, snapshot 2026-10-20.