SaaS-Bench - Software Engineering and Project Management: leaderboard
SaaS-Bench (2026) evaluates computer-use agents on long professional workflows across 23 self-hosted SaaS systems, scoring weighted verification checkpoints on the final application state. This board is the checkpoint score in percent on the software engineering and project management domain; higher is better.
Metric: Checkpoint score (%) on the software engineering and project management 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 February 2027. 14 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Opus 4.7 | 52.9 | #85 | |
| 2 | GPT-5.5 | High | 45.8 | #54 |
| 3 | Claude Opus 4.6 | 41.9 | #113 | |
| 4 | GPT-5.4 | High | 35.5 | #135 |
| 5 | Kimi K2.6 | 25.5 | #192 | |
| 6 | Claude Sonnet 4.6 | 23.9 | #173 | |
| 7 | Seed 2.0 Pro | 22.1 | #222 | |
| 8 | Kimi K2.5 | 20.8 | #266 | |
| 9 | DeepSeek V4 Pro | 20.7 | #196 | |
| 10 | Qwen 3.6 Plus | 20.7 | #238 | |
| 11 | Gemini 3.5 Flash | High | 17.6 | #106 |
| 12 | Gemini 3.1 Pro (Preview) | 14.7 | #99 | |
| 13 | MiniMax-M2.7 | 13.6 | #539 | |
| 14 | GLM-5.1 | 8.7 | #198 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=saas-bench-software-engineering-and-project-management · How It Works · Data refreshed daily, snapshot 2026-11-21.