TKFQA (Table-Text-KG Order): leaderboard
Metric: Exact match (%; final answers to counterfactual multi-hop questions grounded in a table, text passages and a knowledge graph, 1,012 test items; contexts given in table, text, knowledge-graph order; mean of three runs). Source: arxiv.org. Saturation forecast: Around December 2026. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Flash | 84.66 |
| 2 | O3 | 84.5 |
| 3 | Gemini 2.5 Flash | 84.2 |
| 4 | DeepSeek V3.2 | 83.05 |
| 5 | O4 Mini | 82.5 |
| 6 | Kimi K2.5 | 79.31 |
| 7 | Grok 4.3 | 79.22 |
| 8 | GPT-5 | 75.38 |
| 9 | MiniMax-M2.7 | 72.42 |
| 10 | GPT-4.1 Mini | 71.44 |
| 11 | Gemini 2.5 Flash Lite | 69.93 |
| 12 | Qwen 3 30B A3B | 64.9 |
| 13 | Gemma 3 12B | 62.33 |
| 14 | Qwen 3 8B | 52.35 |
Interactive version: theaggregate.ai/benchmark?slug=tkfqa-table-text-kg-order · How It Works · Data refreshed daily, snapshot 2026-09-26.