Finance Agent v2 — leaderboard
Evaluating agents on core financial analyst tasks using the FAB v2 harness.
Metric: Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 33 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.5 Flash | 57.9 |
| 2 | Gemini 3.5 Flash (High) | 57.86 |
| 3 | Claude Fable 5 (Max) | 56.31 |
| 4 | Claude Opus 4.8 (Max) | 53.92 |
| 5 | Claude Opus 4.8 | 53.9 |
| 6 | GPT-5.5 | 51.8 |
| 7 | GPT-5.5 (xHigh) | 51.76 |
| 8 | Claude Opus 4.7 (High) | 51.51 |
| 9 | Claude Opus 4.7 | 51.5 |
| 10 | Claude Sonnet 4.6 (Max) | 51.03 |
| 11 | Qwen 3.7 Max (Max) | 48.35 |
| 12 | MiniMax-M3 | 48.27 |
| 13 | GPT-5.4 Mini (xHigh) | 45.36 |
| 14 | GLM-5.1 | 44.79 |
| 15 | DeepSeek V4 Pro (Max) | 44.08 |
Interactive version: theaggregate.ai/benchmark?slug=finance-agent-v2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.