TinyQABenchmark (core-en): leaderboard
Compact English question-answering probe (TinyQABenchmark core-en) for fast sanity-checking of factual QA capability.
Metric: Exact Match. Source: huggingface.co. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 3 4B | 86.5 | #1089 |
| 2 | Llama 3.2 3B Instruct | 84.6 | #1352 |
| 3 | Ministral 8B | 80.8 | #1040 |
| 4 | Ministral 3B | 76.9 | #1278 |
| 5 | Llama 3.2 1B Instruct | 53.8 | #1539 |
| 6 | Mistral 7B Instruct | 50 | #1363 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=tinyqabenchmark-core-en · How It Works · Data refreshed daily, snapshot 2026-10-20.