# Claude Opus 5.5 finance AI benchmark results

Every finance benchmark score for Claude Opus 5.5 by Anthropic: 6 benchmarks, each with its rank, metric and source date. Scores are never combined across benchmarks.

## Board standings

Claude Opus 5.5 is first on 1 benchmark, top 3 on 4 benchmarks, scored on 6 benchmarks. Places on each benchmark's main board with 2 or more models; scores are never combined. https://benchmarks.finance/method/#standings
- Finance Agent Benchmark: #1 of 10
- FinanceBench: #3 of 5
- Ramp Accounting Bench: #2 of 5
- Vals Finance Index: #2 of 10

## Scores

| Benchmark | Rank | Score | Metric | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [Finance Agent Benchmark](https://benchmarks.finance/benchmarks/vals-finance-agent/) | 1 of 10 | 67.8 | Partial Credit accuracy % (v2.1) | Vals.ai leaderboard | 29 Oct 2026 |
| [Vals Finance Index](https://benchmarks.finance/benchmarks/vals-finance-index/) | 2 of 10 | 65.6 | index accuracy % | Vals.ai leaderboard | 29 Oct 2026 |
| [Excel Modeling Benchmark (EMB)](https://benchmarks.finance/benchmarks/vals-emb/) | 4 of 10 | 75.8 | accuracy % (v1.1) | Vals.ai leaderboard | 29 Oct 2026 |
| [Ramp Accounting Bench](https://benchmarks.finance/benchmarks/ramp-accounting-bench/) | 2 of 5 | 50.5 | mean reward % (criteria, mean of 3 attempts) | Ramp Labs Accounting Bench | 21 Sept 2026 |
| [Tax Agent Bench](https://benchmarks.finance/benchmarks/vals-tax-agent-bench/) | 4 of 10 | 45.6 | accuracy % (all-pass x citation factor) | Vals.ai leaderboard | 29 Oct 2026 |
| [FinanceBench](https://benchmarks.finance/benchmarks/financebench/) | 3 of 5 | 60.3 | % correct, no filing given (150-question sample) | Moneda run, 2026-11-02 | 2 Nov 2026 |

Source: https://benchmarks.finance/models/claude-opus-5-5/

Created and maintained by Moneda.