K-FinHallu - Hallucination Type Classification: leaderboard
Metric: Macro-F1 (0-100) over five hallucination types (contradictory, unverifiable, irrelevance, false acceptance, false refusal) on the 404 hallucinated test responses, multi-turn Korean financial RAG dialogues, temperature 0 where applicable; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 72.5 |
| 2 | GPT-5 | 71.7 |
| 3 | Gemini 2.5 Flash | 66.9 |
| 4 | Qwen 3 32B | 49.2 |
| 5 | Qwen 3 14B | 44.2 |
| 6 | Qwen 3 8B | 38.7 |
| 7 | GPT-4o | 28.9 |
| 8 | Llama 3.1 70B | 13.3 |
| 9 | Llama 3.3 70B | 10.3 |
| 10 | Llama 3.1 8B | 8.7 |
Interactive version: theaggregate.ai/benchmark?slug=k-finhallu-hallucination-type-classification · How It Works · Data refreshed daily, snapshot 2026-10-07.