Bug Hunt Bench - LMS: leaderboard
Coding models find and repair 105 planted defects in two production repositories: 45 in a VS Code extension and 60 in a learning-management system. Blind grading counts verified planted fixes. Reasoning effort, the July DeepSeek revision and the retired Gemini CLI protocol remain distinguishable; superseded runs follow the publisher’s replacement markers. This board reports Bug Hunt Bench - LMS using Planted Bugs Fixed (out of 60).
Metric: Planted Bugs Fixed (out of 60). Source: bughunt.productcompass.pm. Saturation forecast: Around 2028. 132 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 5.5 (Max) | 28.7 | #14 (Claude Sonnet 5.5) |
| 2 | Gemini 4 Argon (High) | 27.7 | #9 (Gemini 4 Argon) |
| 3 | Gemini 4 Argon (Medium) | 25.7 | #9 (Gemini 4 Argon) |
| 4 | GPT-5.6 Sol (Max) | 24.5 | #22 (GPT-5.6 Sol) |
| 5 | DeepSeek V4.1 Flash (1017) (Max) | 24.4 | #27 (DeepSeek V4.1 Flash (1017)) |
| 6 | DeepSeek V4.1 Flash (1017) (High) | 24.4 | #27 (DeepSeek V4.1 Flash (1017)) |
| 7 | Claude Fable 5.1 (Max) | 24 | #10 (Claude Fable 5.1) |
| 8 | GPT-6 Astra (Max) | 23 | #6 (GPT-6 Astra) |
| 9 | Claude Opus 5.5 (Max) | 22.7 | #3 (Claude Opus 5.5) |
| 10 | GPT-6.1 Terra (Max) | 22.7 | #15 (GPT-6.1 Terra) |
| 11 | Gemini 4 Argon (Low) | 22.7 | #9 (Gemini 4 Argon) |
| 12 | GPT-6.1 Sol (Max) | 22 | #7 (GPT-6.1 Sol) |
| 13 | GPT-5.6 Sol (xHigh) | 21 | #22 (GPT-5.6 Sol) |
| 14 | GPT-5.6 Sol (High) | 21 | #22 (GPT-5.6 Sol) |
| 15 | Muse Wonder Full (Max) | 20.5 | #12 (Muse Wonder Full) |
Interactive version: theaggregate.ai/benchmark?slug=bug-hunt-bench-lms · How It Works · Data refreshed daily, snapshot 2026-11-05.