VibeSearchBench (ReAct): leaderboard
Metric: Triplet F1 (%; an LLM judge matches the knowledge graph the agent extracts after a multi-turn session with a progressive-disclosure user simulator against the annotated ground-truth graph; ReAct loop with self-summarising context compaction; mean of VibeSearch-Pro and VibeSearch-Daily; mean of 3 runs per task). Source: arxiv.org. Saturation forecast: Around 2029. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 27.87 |
| 2 | DeepSeek V4 Pro | 27.04 |
| 3 | Kimi K2.6 | 26.09 |
| 4 | Gemini 3.1 Pro (Preview) | 23.54 |
| 5 | Seed 2.0 Pro | 23.22 |
| 6 | Qwen 3.5 397B A17B | 22.84 |
| 7 | GPT-5.4 | 20.04 |
Interactive version: theaggregate.ai/benchmark?slug=vibesearchbench-react · How It Works · Data refreshed daily, snapshot 2026-09-26.