ScarfBench Leaderboard
Comprehensive benchmarking of applications framework migration
Total Submissions
4
Top Overall
66.7
Avg Translate@k
0.0
High (≥80)
Medium (60-79)
Low (<60)
Score Type:
Layers:
From:
To:
Status:
| Rank | Solution | Org | Date | Status |
Layer
|
From | All | Jakarta | Quarkus | Spring | To | All | Jakarta | Quarkus | Spring | Submission |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| F→T | Compile@k | Run@k | Translate@k | |||||||||||||
| #1 | gemini-2.5-pro | Google DeepMind | 2025-12-17 | Computed | Whole App | A→A | 66.7 | 34.4 | - | 🔗 | ||||||
| #2 | claude-sonnet-4.5 | Anthropic | 2025-12-17 | Computed | Whole App | A→A | 53.4 | 21.1 | - | 🔗 | ||||||
| #3 | gpt-5 | OpenAI | 2025-12-17 | Computed | Whole App | A→A | 42.2 | 18.9 | - | 🔗 | ||||||
| #4 | qwen3-coder-480b | Alibaba | 2025-12-17 | Computed | Whole App | A→A | 10.0 | 4.5 | - | 🔗 | ||||||