TRACK - Code Generation (Facts in Context): leaderboard
Metric: Holistic pass (%; BigCodeBench-derived coding problems that require specific library API versions, 500 items; the final answer is correct and the reasoning entails every required fact; answered with the model's own missing facts appended to the prompt). Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 Mini | 36.8 |
| 2 | Qwen 3 4B | 28.2 |
| 3 | Qwen 3 8B | 27.8 |
| 4 | Qwen 3 1.7B | 18.3 |
Interactive version: theaggregate.ai/benchmark?slug=track-code-generation-facts-in-context · How It Works · Data refreshed daily, snapshot 2026-09-26.