FinSkillBench (Self-Generated Skills): leaderboard
Metric: Task score (0-1, native; the agent writes and reuses its own skill documents within each episode; over FinSkillBench's 12 investment-management subtasks (portfolio construction, risk management, fundamental analysis), unweighted mean, hidden ground truth and task-specific verifiers). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 0.49 |
| 2 | DeepSeek V3.2 | 0.45 |
| 3 | Gemini 3.1 Flash Lite | 0.43 |
| 4 | Gemini 2.5 Pro | 0.41 |
| 5 | GPT-5.4 | 0.41 |
| 6 | GPT-4.1 | 0.4 |
| 7 | GLM-5.1 | 0.25 |
| 8 | Grok 4 | 0.14 |
| 9 | Phi-4 | 0 |
Interactive version: theaggregate.ai/benchmark?slug=finskillbench-self-generated-skills · How It Works · Data refreshed daily, snapshot 2026-09-26.