SaaS-Bench - Healthcare Administration: leaderboard
Metric: Checkpoint score (%) on the healthcare administration domain, weighted-checkpoint score over the final application states, browser-use agent loop over live self-hosted SaaS apps (DOM and screenshot observations, restricted browser actions, no JavaScript or backend access), same step budget, timeout and failure cap for every model; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (High) | 30.9 |
| 2 | Claude Opus 4.7 | 29.7 |
| 3 | Claude Opus 4.6 | 27.2 |
| 4 | GPT-5.4 (High) | 26.8 |
| 5 | Kimi K2.6 | 26.2 |
| 6 | GLM-5.1 | 24 |
| 7 | Kimi K2.5 | 20.2 |
| 8 | Gemini 3.1 Pro (Preview) | 20 |
| 9 | Qwen 3.6 Plus | 18.9 |
| 10 | MiniMax-M2.7 | 17.7 |
| 11 | DeepSeek V4 Pro | 17.1 |
| 12 | Gemini 3.5 Flash (High) | 14.5 |
| 13 | Seed 2.0 Pro | 13.6 |
| 14 | Claude Sonnet 4.6 | 9.5 |
Interactive version: theaggregate.ai/benchmark?slug=saas-bench-healthcare-administration · How It Works · Data refreshed daily, snapshot 2026-10-07.