MM-IssueLoc - Function Localization: leaderboard
MM-IssueLoc tests whether repository-level localization systems use screenshots and other issue images to identify the functions responsible for a software issue; scored by function Acc@10 with the best harness retained per underlying LLM.
Metric: Function Acc@10 (%). Source: arxiv.org. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 22.45 | #110 |
| 2 | Claude Sonnet 4.6 | 20.41 | #90 |
| 3 | Kimi K2.5 | 20.12 | #144 |
| 4 | Qwen 3.5 122B A10B | 12.98 | #173 |
| 5 | Gemini 3 Pro | 12.92 | #80 |
Interactive version: theaggregate.ai/benchmark?slug=mm-issueloc-function-localization · How It Works · Data refreshed daily, snapshot 2026-10-20.