EconLogicQA: leaderboard
Multi-step economic-reasoning benchmark asking models to order and justify sequences of economic events and decisions.
Metric: Accuracy. Source: huggingface.co. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 Turbo | 0.57 | #404 |
| 2 | GPT-4 | 0.55 | #438 |
| 3 | GPT-3.5 Turbo | 0.38 | #865 |
| 4 | Llama 3 8B Instruct | 0.35 | #1129 |
| 5 | Mistral 7B Instruct (v0.2) | 0.32 | #1333 |
| 6 | Mistral-7B-v0.1 | 0.26 | #1326 |
| 7 | Mistral-7B-v0.2 | 0.26 | #1388 |
| 8 | Llama 3 8B | 0.24 | #1242 |
| 9 | zephyr-7B-alpha | 0.23 | #1312 |
| 10 | Yi 6B (Chat) | 0.21 | #1408 |
| 11 | zephyr-7B-beta | 0.18 | #1330 |
| 12 | Mistral Instruct v0.1 7B | 0.15 | #1454 |
| 13 | Llama 2 13B Chat | 0.15 | #1465 |
| 14 | Llama 2 7B Chat | 0.09 | #1499 |
| 15 | gemma-2B (IT) | 0.08 | #1559 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=econlogicqa · How It Works · Data refreshed daily, snapshot 2026-10-20.