APEX-Agents — leaderboard
The AI Productivity Index for Agents (APEX-Agents) measures whether frontier AI agents can execute long-horizon, cross-application tasks across three jobs in professional services.
Metric: Mean Score (ReAct) (self-reported). Source: benchmarklist.com. Status: saturation imminent. 40 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.5 Flash | 66.1 |
| 2 | Claude Opus 4.8 | 59.4 |
| 3 | GPT-5.5 (xHigh) | 53.9 |
| 4 | GPT-5.4 (xHigh) | 52.7 |
| 5 | Claude Opus 4.7 | 50.6 |
| 6 | GPT-5.2 (xHigh) | 48.4 |
| 7 | Claude Opus 4.6 (Max) | 48.4 |
| 8 | Gemini 3.1 Pro (Preview) (High) | 48.2 |
| 9 | GPT-5.3 Codex (High) | 46.9 |
| 10 | GPT-5.2 Codex (High) | 42.2 |
| 11 | Claude Sonnet 4.6 (High) | 40.7 |
| 12 | Gemini 3 Flash (Preview) (High) | 39.5 |
| 13 | GPT-5.4 Mini (xHigh) | 37.5 |
| 14 | GPT-5.1 Codex (High) | 34.9 |
| 15 | GPT-5 Codex (High) | 34.8 |
Interactive version: theaggregate.ai/benchmark?slug=apex-agents · How the rankings work · Data refreshed daily, snapshot 2026-07-22.