Z.ai GLM-5.3 Launch - SWE-Marathon v1.1: leaderboard
SWE-Marathon v1.1 ultra-long-horizon software-engineering tasks as reported in Z.ai's GLM-5.3 launch table; Z.ai-run frontier comparison using Claude Code at maximum effort.
Metric: Pass@1 (%). Source: z.ai. Saturation forecast: Around January 2028. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.8 | 48.8 | #37 |
| 2 | Kimi K3 | 48.1 | #30 |
| 3 | GPT-5.6 Sol | 42.5 | #19 |
| 4 | GLM-5.3 | 42.5 | #39 |
| 5 | Claude Fable 5 | 33.1 | #16 |
| 6 | GLM-5.2 | 19.4 | #73 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=z-ai-glm-5-3-launch-swe-marathon-v1-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.