Benchmarks · Coding · Agents & computer use
LongTail-SWE: Kotlin
LongTail-SWE's Kotlin tasks: fix real GitHub issues in Kotlin projects. The share of attempts that pass the tests.
11 models · 11 results · data as of 2026-11-18 · longtail-swe.com ↗
Leaderboard
Opus 5.5 high
All results
11 rows| # | Model | Effort | LongTail-SWE: Kotlin | Harness |
|---|---|---|---|---|
| 1 | Opus 5.5 | high | 77.8% | SWE-agent-like scaffold |
| 2 | Kimi K3 | max | 75.0% | SWE-agent-like scaffold |
| 3 | GPT-5.6 Sol | high | 72.2% | SWE-agent-like scaffold |
| 4 | Kimi K3.1 | max | 72.2% | SWE-agent-like scaffold |
| 5 | GPT-6.1 Sol | high | 69.4% | SWE-agent-like scaffold |
| 6 | DeepSeek V4 Flash | high | 66.7% | SWE-agent-like scaffold |
| 7 | DeepSeek V4.1 Flash 1017 | max | 66.7% | SWE-agent-like scaffold |
| 8 | GPT-5.6 Luna | xhigh | 61.1% | SWE-agent-like scaffold |
| 9 | GLM 5.3 Flash | max | 58.3% | SWE-agent-like scaffold |
| 10 | Qwen 3.8 27B | xhigh | 47.2% | SWE-agent-like scaffold |
| 11 | Qwen 3.6 27B | max | 22.2% | SWE-agent-like scaffold |