DS-1000: leaderboard
DS-1000: Measures model capability on programming, code generation, code repair, or repository-level software tasks.
Metric: Overall mean. Source: github.com. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 Turbo | 0.54 | #404 |
| 2 | GPT-4 (0613) | 0.51 | #491 |
| 3 | GPT-3.5 Turbo (0125) | 0.39 | #984 |
| 4 | GPT-3.5 Turbo (0613) | 0.39 | #1118 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=ds-1000 · How It Works · Data refreshed daily, snapshot 2026-10-20.