Agentick — leaderboard

Universal benchmark for evaluating sequential decision agents across RL, LLM, VLM, modality, harness, task, and category settings.

Metric: Oracle-Normalized Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 10 models tracked.

Top models

#ModelScore
1GPT-5 Mini30.9
2Gemini 2.5 Flash Lite18.7
3Qwen 3.5 2B13.3
4Qwen 3.5 0.8B9.4
5Qwen 3 4B8.5

Interactive version: theaggregate.ai/benchmark?slug=agentick · How the rankings work · Data refreshed daily, snapshot 2026-07-22.