# Finance Agent Benchmark leaderboard

Checks whether a tool-using agent can do entry-level financial analyst research, such as answering hard questions from public company filings. Question types include quantitative and qualitative analysis, comparables, adjustments, earnings, disclosures and modelling.

## Finance Agent Benchmark: Partial Credit accuracy % (higher is better, 11 models)

| Rank | Model | Maker | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | Numa 1 (public set) | Moneda | 80.2 | Partial Credit accuracy % | Public set (27 questions), run by Moneda with its own grader (GPT-4o per rubric criterion); not a Vals result. Vals' other rows are its test set. One question is worth about 4 points. | Run by Moneda | Moneda run, 2026-10-09 | https://benchmarks.finance/method/ | 9 Oct 2026 |
| 2 | Claude Opus 5.5 | Anthropic | 67.8 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 3 | Gemini 4 Argon | Google | 67.7 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 4 | Wonder Full | Not stated | 67.6 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 5 | Claude Sonnet 5.5 | Anthropic | 67.3 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 6 | Gemini 4 Carbon | Google | 67.2 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 7 | Gemini 3.8 Flash | Google | 65.8 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 8 | Wonder | Not stated | 65.5 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 9 | DeepSeek V4.1 Flash | DeepSeek | 65.4 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 10 | Claude Haiku 5.5 | Anthropic | 65.4 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |
| 11 | GLM 5.3 | Zhipu AI | 65 | Partial Credit accuracy % | v2.1 test set | Official board | Vals.ai leaderboard | https://www.vals.ai/benchmarks/fabv2-1 | 29 Oct 2026 |

Scores are never combined across benchmarks. Each row names its metric, source and date.

## About this benchmark

- Maker: Vals AI
- Year: 2025
- Use case: Analyst Financial research
- Size: v2.1: 926 questions (27 public, 450 private validation, 449 test). v1 paper: 537 questions.
- Public split: 27 public questions plus the agent harness in the GitHub repo. The v1 paper released 50 public validation questions (CC BY 4.0).
- Grading: Each question has weighted rubric checks, some marked as dealbreakers. A missed dealbreaker scores the question zero. Headline metric is Partial Credit; All-Pass is the secondary. A three-model LLM jury grades (GPT-6.1 Sol, Gemini 3.8 Flash, Claude Sonnet 5.5). Two-hour limit per task.
- Licence: MIT (code, finance-agent-v2 repo). The v1 paper lists CC BY 4.0 for the public data and MIT for the code.
- Submission: Not stated
- Products allowed: Not stated
- Latest version: v2.1
- What it does not cover: Does not test trading, portfolio decisions, advice suitability or spreadsheet model building.
- Last checked: 1 Nov 2026
- Links: paper https://arxiv.org/abs/2508.00828, site https://www.vals.ai/benchmarks/fabv2-1, code https://github.com/vals-ai/finance-agent-v2, leaderboard https://www.vals.ai/benchmarks/fabv2-1
- Directory page: https://benchmarks.finance/benchmarks/vals-finance-agent/
- Leaderboard: https://benchmarks.finance/leaderboards/vals-finance-agent/

Source: https://benchmarks.finance/leaderboards/vals-finance-agent/

Created and maintained by Moneda.