K-FinHallu - Four-Class Detection: leaderboard
Accuracy (%) at labelling each dialogue response as faithful, hallucinated, a false refusal or a true refusal given the retrieved passages in K-FinHallu (2026), a benchmark of multi-turn Korean financial RAG dialogues built from public financial documents, with injected hallucinations and refusal cases. Higher is better.
Metric: Overall accuracy (%), micro-averaged over 808 test responses, at labelling each response faithful, hallucinated, false refusal or true refusal given the retrieved passages, multi-turn Korean financial RAG dialogues, temperature 0 where applicable; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 85.5 | #178 |
| 2 | Gemini 2.5 Pro | 85.3 | #276 |
| 3 | Gemini 2.5 Flash | 84.5 | #463 |
| 4 | Qwen 3 32B | 74.3 | #972 |
| 5 | Qwen 3 14B | 74.1 | #1269 |
| 6 | Llama 3.1 70B | 69.3 | #1499 |
| 7 | GPT-4o | 69.2 | #751 |
| 8 | Qwen 3 8B | 68.8 | #1873 |
| 9 | Llama 3.3 70B | 68.7 | #1514 |
| 10 | EXAONE 4.0 32B | 60.1 | #1398 |
| 11 | kanana-2-30B-a3B-instruct | 51.5 | |
| 12 | Llama 3.1 8B | 34.2 | #5196 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=k-finhallu-four-class-detection · How It Works · Data refreshed daily, snapshot 2026-11-21.