SearchAgent - TriviaQA — leaderboard
Metric: Exact Match (%). Source: huggingface.co. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | ZeroSearch-Qwen2.5-7b-Instruct | 65.2 |
| 2 | ReSeek-Qwen2.5-7b-Instruct | 64 |
| 3 | Search-R1-Qwen2.5-7b-Instruct | 61 |
| 4 | RAG-Qwen2.5-7b-Instruct | 58.5 |
| 5 | ZeroSearch-Qwen2.5-3b-Instruct | 57.4 |
| 6 | ReSeek-Qwen2.5-3b-Instruct | 55.3 |
| 7 | Search-R1-Qwen2.5-3b-Instruct | 54.5 |
| 8 | R1-Qwen2.5-7b-Instruct | 53.7 |
| 9 | Search-o1-Qwen2.5-7b-Instruct | 44.3 |
| 10 | Direct-Inference-Qwen2.5-7b-Instruct | 40.8 |
Interactive version: theaggregate.ai/benchmark?slug=searchagent-triviaqa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.