Exgentic Open Agent - TauBench-Retail: leaderboard
Exgentic Open Agent TauBench-Retail evaluates tool-using agents in retail customer-service workflows.
Metric: Success Rate (%). Source: www.exgentic.ai. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | OpenAI Solo + claude-opus-4.5 | 85 | |
| 2 | Claude Code + claude-opus-4.5 | 83 | |
| 3 | React + deepseek-v3.2 | 82 | |
| 4 | React + gemini-3-pro | 82 | |
| 5 | React + Shortlisting + deepseek-v3.2 | 82 | |
| 6 | React + Shortlisting + gemini-3-pro | 82 | |
| 7 | Smolagent + claude-opus-4.5 | 78 | |
| 8 | React + claude-opus-4.5 | 78 | |
| 9 | React + Shortlisting + claude-opus-4.5 | 78 | |
| 10 | Smolagent + deepseek-v3.2 | 77 |
Interactive version: theaggregate.ai/benchmark?slug=exgentic-open-agent-taubench-retail · How It Works · Data refreshed daily, snapshot 2026-10-20.