Finance Agent v2: leaderboard
Evaluating agents on core financial analyst tasks using the FAB v2 harness.
Metric: Accuracy (%). Source: vals.ai. Source: benchmarklist.com. 80 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon | 65.4 | #8 |
| 2 | Gemini 3.8 Flash | 61.44 | #21 |
| 3 | Muse Spark Open | 61.34 | #54 |
| 4 | Muse Spark 1.2 | 60.6 | #62 |
| 5 | Muse Spark 1.3 (Max) | 59.96 | #24 (Muse Spark 1.3) |
| 6 | Muse Wonder Full (Max) | 59.1 | #10 (Muse Wonder Full) |
| 7 | Gemini 3.7 Flash | 59.04 | #23 |
| 8 | Muse Spark 1.3 | 58.9 | #24 |
| 9 | Claude Fable 5.1 | 58.88 | #9 |
| 10 | Claude Opus 5 | 58.63 | #12 |
| 11 | Claude Opus 5.5 | 58.59 | #3 |
| 12 | Claude Sonnet 5.5 | 58.1 | #11 |
| 13 | Ling 3.1 Flash | 57.87 | #71 |
| 14 | Gemini 3.5 Flash (High) | 57.86 | #60 (Gemini 3.5 Flash) |
| 15 | GLM-5.3 Flash | 57.85 | #65 |
Interactive version: theaggregate.ai/benchmark?slug=finance-agent-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.