DistDebug-Bench - Mention: leaderboard
DistDebug Table 9:120 common incidents,three selected clean attempts per bug,mean within bug then equal across bugs; native file/function/line/causal-block/mention localization and block precision. Flagged answer-leaking attempts discarded and rerun; no best-attempt selection. Explicit medium effort for GPT and Claude; source says medium where controls exist and provider defaults otherwise. Model/harness configurations stay separate. These are root-cause localization metrics, not patch success. Metric: Coverage (%).
Metric: Coverage (%). Source: arxiv.org. Saturation forecast: Around March 2027. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Medium) | 82.8 | #6 (GPT-6 Astra) |
| 2 | Claude Opus 5.5 (Medium) | 80.6 | #3 (Claude Opus 5.5) |
| 3 | GPT-5.5 (Medium) | 74.7 | #33 (GPT-5.5) |
| 4 | MiMo-V2.5-Pro | 72.6 | #167 |
| 5 | Kimi K2.6 | 70.6 | #113 |
| 6 | MiMo-V2.5 | 69.5 | #196 |
| 7 | GPT-5.4 (Medium) | 69 | #83 (GPT-5.4) |
| 8 | GPT-5.3 Codex (Medium) | 68.9 | #72 (GPT-5.3 Codex) |
| 9 | DeepSeek V4 Flash | 67 | #155 |
| 10 | Hy3 | 64.5 | #100 |
| 11 | Gemini 3.5 Flash | 58.5 | #66 |
| 12 | O4 Mini (Medium) | 49.8 | #185 (O4 Mini) |
No result here: #7 GPT-6.1 Sol, #9 Gemini 4 Argon, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=distdebug-bench-mention · How It Works · Data refreshed daily, snapshot 2026-11-05.