Best Open-Source Multi-Modal Agent AI Agents (2026)
As of June 2026, the top open-source multi-modal agent AI agent is smolagents by huggingface, with an overall score of 85.8 / 100. 4 multi-modal agent agents are tracked in this category on The Agentic Leaderboard.
| Rank | Agent | Score |
|---|---|---|
| #14 | smolagents by huggingface | 85.8 |
| #16 | ppt-master by hugohe3 | 85.7 |
| #64 | hyperframes by heygen-com | 83.5 |
| #67 | UI-TARS-desktop by bytedance | 83.4 |