ReliableMath - Prudence — leaderboard

Metric: Score (%). Source: huggingface.co. 20 models tracked.

Top models

#ModelScore
1GPT-4o (2024-08-06)1.5
2doubao-seed-1.6-thinking-2506151
3O3 Mini (2025-01-31)0.6
4DeepSeek R10.4
5Qwen 3 14B0.2
6DeepSeek R1 Distill Qwen 32B0.1
7DeepSeek V30.1
8Qwen 3 235B A22B0.1
9Gemini 2.5 Flash (Preview 04-17)0.1
10Gemini 2.5 Flash (Preview 04-17) (Thinking)0.1
11Qwen 3 32B0
12DeepSeek R1 Distill Qwen 14B0
13DeepSeek R1 05280
14DeepSeek-R1-Distill-Qwen-7B0
15DeepSeek R1 Distill Qwen 1.5B0

Interactive version: theaggregate.ai/benchmark?slug=reliablemath-prudence · How the rankings work · Data refreshed daily, snapshot 2026-07-22.