Skip to content

SWE-bench Verified (mini-swe-agent-v2)

A display-only SWE-bench Verified reference from Arcee AI's Trinity-Large-Thinking comparison chart.

5 models

Higher is better.

  1. 1Claude Opus 4.6Anthropic75.6
  2. 2MiniMax M2.7MiniMax75.4
  3. 3GLM-5Z.AI72.8
  4. 4Kimi K2.5Moonshot AI70.8
  5. 5Trinity-Large-ThinkingArcee AI63.2
Higher is better. Bar colour: blue proprietary, orange open weights, green other.

About this benchmark

Metric
Agent scaffold benchmark
Tasks
Repository task completion
Authors
Arcee AI (2026)

Scores are as reported by model developers or evaluators and may use different settings (for example reasoning effort or tools). Compare with care.