Flame Personal Finance - Evaluate: leaderboard
Metric: Accuracy (%; the paper's Bloom-level weighted accuracy on the Evaluate problems of the 199-problem textbook-grounded Personal Finance multiple-choice set, zero-shot, temperature 0). Source: arxiv.org. Saturation forecast: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 100 |
| 2 | Claude Opus 4.6 | 96 |
| 3 | GPT-4.1 Mini | 88 |
| 4 | GPT-5.4 | 81 |
| 5 | Claude Haiku 4.5 | 77 |
| 6 | Gemini 2.5 Flash Lite | 75 |
| 7 | Qwen 3 32B | 58 |
| 8 | DeepSeek R1 Distill Qwen 32B | 52 |
| 9 | Gemma 3 27B (IT) | 42 |
| 10 | Qwen 3 8B | 40 |
| 11 | DeepSeek R1 Distill Qwen 14B | 40 |
| 12 | Gemma 3 12B (IT) | 35 |
Interactive version: theaggregate.ai/benchmark?slug=flame-personal-finance-evaluate · How It Works · Data refreshed daily, snapshot 2026-09-26.