FIND: leaderboard
Financial decision-making in multilingual settings demands accurate numerical reasoning grounded in diverse modalities, yet existing benchmarks largely overlook this high-stakes, real-world challenge, especially for Indic languages. We introduce FinVQA, a benchmark for evaluating financial numerical and multimodal reasoning in multilingual Indic contexts. FinVQA spans English, Hindi, Bengali, Mara
Metric: Zero shot en (self-reported). Source: benchmarklist.com. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 32B Instruct | 77.8 | #283 |
| 2 | Qwen 3 VL 8B Instruct | 74.1 | #420 |
| 3 | Qwen 2.5 VL 32B Instruct | 73.8 | #453 |
| 4 | Gemma 3 27B (IT) | 73.7 | #523 |
| 5 | Gemma 3 12B (IT) | 71.7 | #664 |
| 6 | Qwen 3 VL 4B Instruct | 69.6 | #515 |
| 7 | Gemma 3 4B (IT) | 61.5 | #992 |
| 8 | Qwen 2.5 VL 7B Instruct | 55.6 | #654 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=find · How It Works · Data refreshed daily, snapshot 2026-10-20.