PrismML Bonsai 2 27B - XHigh - SWE-bench Verified (mini-swe-agent): leaderboard
PrismML compares Bonsai 2 27B and Qwen3.8 27B at xhigh reasoning effort on 500 software-issue tasks using mini-swe-agent. Resolved rate (%).
Metric: Resolved rate (%). Source: raw.githubusercontent.com. Saturation forecast: Around January 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.8 27B (xHigh) | 80.6 | #116 (Qwen 3.8 27B) |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=prismml-bonsai-2-27b-xhigh-swe-bench-verified-mini-swe-agent · How It Works · Data refreshed daily, snapshot 2026-10-29.