Grip on LLMs - HonestCityBench - Non-Text Modality: leaderboard
Metric: Appropriate acknowledgement rate (%; LLM judge). Source: arxiv.org. 31 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 51 |
| 2 | Gemma 3 12B (IT) | 38 |
| 3 | Phi-4 Mini Instruct | 37 |
| 4 | GPT-4o Mini | 35 |
| 5 | Gemma 3 27B (IT) | 31 |
| 6 | Qwen 3 32B | 24 |
| 7 | OLMo-2-1124-7B-Instruct | 24 |
| 8 | DeepSeek R1 Distill Qwen 32B | 20 |
| 9 | Mistral Small 3 | 20 |
| 10 | GEITje-7B-ultra | 18 |
| 11 | Llama 3.1 8B Instruct | 15 |
| 12 | command-r7B-12-2024 | 15 |
| 13 | aya-expanse-32B | 14 |
| 14 | Mistral Medium 3 | 13 |
| 15 | GPT-OSS-20B | 10 |
Interactive version: theaggregate.ai/benchmark?slug=grip-on-llms-honestcitybench-non-text-modality · How It Works · Data refreshed daily, snapshot 2026-09-19.