GuardianAgentBench (LlamaIndex): leaderboard

Metric: Overall correctness (%; mean over six agent domains of the share of scenarios whose answer and tool-call trace are both judged correct by Claude Sonnet 4.5; 580 scenarios with 81 tools, a third with adversarially modified tool responses; LlamaIndex FunctionAgent, default API settings). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 6 models tracked.

Top models

#ModelScore
1LlamaIndex + Claude Opus 4.573.7
2LlamaIndex + Gemini 3 Pro72.3
3LlamaIndex + GPT-5.2 Pro70.8
4LlamaIndex + Qwen 3 Max66.8
5LlamaIndex + GPT-OSS-120B66.1
6LlamaIndex + DeepSeek V3.264.2

Interactive version: theaggregate.ai/benchmark?slug=guardianagentbench-llamaindex · How It Works · Data refreshed daily, snapshot 2026-09-29.