ArkEval (Official-Knowledge RAG): leaderboard
Metric: Pass@1 (%): share of the 502 instances whose patch compiles and passes the reproduction test, with retrieved official ArkTS documentation added during patch generation, same localization and gates as the RAG-off runs; higher is better. Source: arxiv.org. Saturation forecast: Around August 2028. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol | 20.92 | #16 |
| 2 | MiniMax-M3 | 10.36 | #129 |
| 3 | DeepSeek V4 Pro | 8.17 | #96 |
| 4 | MiniMax-M2.5 | 2.79 | #295 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=arkeval-official-knowledge-rag · How It Works · Data refreshed daily, snapshot 2026-10-11.