Real-World Finance v2: leaderboard
Anthropic long-horizon finance workflow evaluation using pairwise preference grading and Elo ratings over realistic professional deliverables.
Metric: Elo. Source: www-cdn.anthropic.com. Saturation forecast: Around October 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 1374 | #16 |
| 2 | Claude Mythos Preview | 1307 | #4 |
| 3 | Claude Opus 4.8 | 1222 | #37 |
| 4 | Claude Sonnet 4.6 | 1000 | #90 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=real-world-finance-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.