FinanceArena: leaderboard
FinanceArena’s FinanceQA leaderboard for industry-grade financial analysis tasks across basic tactical, assumption-based tactical, and conceptual categories.
Metric: Overall Accuracy (imported). Source: www.afterquery.com. Source: benchmarklist.com. Saturation forecast: Around November 2026. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 54.1 | #124 |
| 2 | Grok 4 | 49.3 | #181 |
| 3 | O4 Mini | 48.6 | #180 |
| 4 | Gemini 2.5 Pro | 45.3 | #152 |
| 5 | Llama 4 Maverick | 44.6 | #464 |
| 6 | Grok 3 | 44.6 | #304 |
| 7 | Claude Opus 4 | 44.6 | #163 |
| 8 | Claude Sonnet 4 | 43.9 | #199 |
| 9 | Phi-4-reasoning-plus (Thinking) | 43.2 | |
| 10 | DeepSeek R1 0528 | 42.9 | #222 |
| 11 | QwQ-32B | 42.6 | #421 |
| 12 | GPT-4.1 Mini | 41.9 | #355 |
| 13 | Llama 3.1 Nemotron Ultra 253B v1 | 41.9 | #441 |
| 14 | Qwen 3 30B A3B | 37.2 | #493 |
| 15 | Kimi K2 | 33.8 | #243 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=financearena · How It Works · Data refreshed daily, snapshot 2026-10-20.