GuardianAgentBench (Vectara): 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; Vectara cloud agent platform, default API settings). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 6 models tracked.

Top models

#ModelScore
1Vectara + Claude Opus 4.574.8
2Vectara + GPT-5.2 Pro72.1
3Vectara + Gemini 3 Pro71.4
4Vectara + Qwen 3 Max68.3
5Vectara + GPT-OSS-120B65.8
6Vectara + DeepSeek V3.263

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