Coding · Benchmark profile
A display-only SWE-bench Verified reference from Arcee AI's Trinity-Large-Thinking comparison chart.
Data verified 21 Jul 2026 · Methodology 1.6.0
Benchmark score on SWE-bench Verified (mini-swe-agent-v2)
Claude Opus 4.6 leads at 75.6%, followed by MiniMax M2.7 (75.4%) and GLM-5 (72.8%).
Visual analysis
Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.
5 ranked models · higher is better
| Model | Rank | Provider | Score |
|---|---|---|---|
| Claude Opus 4.6 | #1 | Anthropic | 75.6% |
| MiniMax M2.7 | #2 | MiniMax | 75.4% |
| GLM-5 | #3 | Z.AI | 72.8% |
| Kimi K2.5 | #4 | Moonshot AI | 70.8% |
| Trinity-Large-Thinking | #5 | Arcee AI | 63.2% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | Claude Opus 4.6 claude-opus-4-6 | Anthropic | closed | Reference only | 75.6% |
| #2 | MiniMax M2.7 minimax-m2-7 | MiniMax | open | Reference only | 75.4% |
| #3 | GLM-5 glm-5 | Z.AI | open | Reference only | 72.8% |
| #4 | Kimi K2.5 kimi-k2.5 | Moonshot AI | — | Reference only | 70.8% |
| #5 | Trinity-Large-Thinking trinity-large-thinking | Arcee AI | open | Reference only | 63.2% |
The top of this snapshot is led by Claude Opus 4.6 at 75.6%; third place is 2.8 points behind. The top-5 spread is 12.4 points.
About SWE-bench Verified (mini-swe-agent-v2)
A display-only SWE-bench Verified reference from Arcee AI's Trinity-Large-Thinking comparison chart. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A display-only SWE-bench Verified reference from Arcee AI's Trinity-Large-Thinking comparison chart.
Claude Opus 4.6 by Anthropic currently leads with 75.6%.
5 models in the LuminaBench cohort have a qualifying score on this benchmark.
No. This benchmark is display-only and does not enter the overall Lumina composite.
Related
