Sonar LLM Leaderboard - Java - Pass Rate: leaderboard
Percentage of unit tests passed on Sonar's LLM Leaderboard, which has each model solve a fixed set of Java programming assignments and analyses the generated code with SonarQube; higher is better.
Metric: Passing tests (%). Source: www.sonarsource.com. Saturation forecast: Around February 2027. 74 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (Thinking) | 88.6 | #16 (Claude Opus 5) |
| 2 | GPT-6 Astra (Medium) | 85.85 | #6 (GPT-6 Astra) |
| 3 | Gemini 4 Argon (High) | 85.66 | #9 (Gemini 4 Argon) |
| 4 | GPT-6.1 Sol (Medium) | 85.29 | #7 (GPT-6.1 Sol) |
| 5 | Gemini 3.1 Pro (Preview) (High) | 84.17 | #64 (Gemini 3.1 Pro (Preview)) |
| 6 | GPT-6 Sol (High) | 83.64 | #19 (GPT-6 Sol) |
| 7 | Claude Opus 4.5 (High) | 83.63 | #87 (Claude Opus 4.5) |
| 8 | GPT-6 Sol (Medium) | 83.09 | #19 (GPT-6 Sol) |
| 9 | Claude Opus 4.8 (High) | 82.9 | #39 (Claude Opus 4.8) |
| 10 | Claude Opus 4.6 (High) | 82.72 | #67 (Claude Opus 4.6) |
| 11 | GPT-5.6 Sol (Medium) | 81.99 | #22 (GPT-5.6 Sol) |
| 12 | Claude Sonnet 4.6 (High) | 81.76 | #94 (Claude Sonnet 4.6) |
| 13 | Claude Opus 4.7 (High) | 81.65 | #51 (Claude Opus 4.7) |
| 14 | Gemini 3 Pro (High) | 81.6 | #82 (Gemini 3 Pro) |
| 15 | Kimi K2.6 | 81.43 | #113 |
No result here: #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=sonar-llm-leaderboard-java-pass-rate · How It Works · Data refreshed daily, snapshot 2026-11-05.