LongDS v1.1 Full - Business: leaderboard
Metric: Accuracy (%; judged turn accuracy macro-averaged over 12 tasks), each model under one agentic coding harness. Source: zjunlp.github.io. Saturation forecast: Around March 2027. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 56.99 |
| 2 | Kimi K3 | 43.35 |
| 3 | Qwen 3.8 Max (0902) | 41.96 |
| 4 | GLM-5.2 | 33.61 |
| 5 | DeepSeek V4 Pro | 16.59 |
Interactive version: theaggregate.ai/benchmark?slug=longds-v1-1-full-business · How It Works · Data refreshed daily, snapshot 2026-09-25.