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

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

Top Models on SWE-bench Verified* — March 2026

As of March 2026, Claude Opus 4.6 leads the SWE-bench Verified* leaderboard with 75.6% , followed by MiniMax M2.7 (75.4%) and GLM-5 (72.8%).

5 modelsCodingCurrentDisplay onlyUpdated April 1, 2026

According to BenchLM.ai, Claude Opus 4.6 leads the SWE-bench Verified* benchmark with a score of 75.6%, followed by MiniMax M2.7 (75.4%) and GLM-5 (72.8%). The top models are clustered within 2.8 points, suggesting this benchmark is nearing saturation for frontier models.

5 models have been evaluated on SWE-bench Verified*. The benchmark falls in the Coding category. This category carries a 20% weight in BenchLM.ai's overall scoring system. SWE-bench Verified* is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About SWE-bench Verified*

Year

2026

Tasks

Repository task completion

Format

Agent scaffold benchmark

Difficulty

Professional software engineering

BenchLM stores this chart-specific SWE-bench Verified row separately because Arcee notes all models were evaluated in mini-swe-agent-v2.

Trinity-Large-Thinking: Scaling an Open Source Frontier Agent

BenchLM freshness & provenance

Version

SWE-bench Verified* 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Leaderboard (5 models)

#1Claude Opus 4.6
75.6%
#2MiniMax M2.7
75.4%
#3GLM-5
72.8%
#4Kimi K2.5
70.8%
#5Trinity-Large-Thinking
63.2%

FAQ

What does SWE-bench Verified* measure?

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

Which model scores highest on SWE-bench Verified*?

Claude Opus 4.6 by Anthropic currently leads with a score of 75.6% on SWE-bench Verified*.

How many models are evaluated on SWE-bench Verified*?

5 AI models have been evaluated on SWE-bench Verified* on BenchLM.

Last updated: April 1, 2026 · BenchLM version SWE-bench Verified* 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.