ScaleDown CUAD - OpenAI Comparison: leaderboard
ScaleDown’s published CUAD comparison measures contract span-extraction F1 (0–1) for its Extract API and GPT-5.5, GPT-5.4 Mini and GPT-5.4 Nano. Higher is better. Prompt, reasoning effort and exact evaluation date are undisclosed; results remain separate from other provider cohorts.
Metric: CUAD span-extraction F1 (0–1). Source: scaledown.ai. Saturation forecast: Around 2029. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 0.5 | #29 |
| 2 | GPT-5.4 Mini | 0.47 | #213 |
| 3 | GPT-5.4 Nano | 0.43 | #329 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=scaledown-cuad-openai-comparison · How It Works · Data refreshed daily, snapshot 2026-10-22.