SaaS-Bench - Text-Only: 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 text-only tasks; higher is better.

Metric: Checkpoint score (%) on the 74 text-only tasks of the four text-only domains, 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 August 2027. 14 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 4.742.8#85
2GPT-5.5High42.1#54
3Claude Opus 4.640.7#113
4GPT-5.4High33#135
5Kimi K2.630.1#192
6Kimi K2.523.6#266
7Qwen 3.6 Plus23.1#238
8DeepSeek V4 Pro21.5#196
9Gemini 3.1 Pro (Preview)20.6#99
10Seed 2.0 Pro19.8#222
11Claude Sonnet 4.618.7#173
12Gemini 3.5 FlashHigh17.6#106
13GLM-5.117.4#198
14MiniMax-M2.715.8#539

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-text-only · How It Works · Data refreshed daily, snapshot 2026-11-21.