reasoning benchmark
Finance Agent v2 is an agentic financial-analysis benchmark from Vals that evaluates models on financial workflows.
Updated Sep 5, 2026
Higher score ranks better on this benchmark.
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelGO | Score57.86% | Percentile100.00% | Participants26 | EvidenceB | Evaluated |
| Rank02 | ModelME | Score57.20% | Percentile96.00% | Participants26 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score56.31% | Percentile92.00% | Participants26 | EvidenceB | Evaluated |
| Rank04 | ModelAN | Score53.92% | Percentile88.00% | Participants26 | EvidenceB | Evaluated |
| Rank05 | ModelOP | Score51.76% | Percentile84.00% | Participants26 | EvidenceB | Evaluated |
| Rank06 | ModelAN | Score51.51% | Percentile80.00% | Participants26 | EvidenceB | Evaluated |
| Rank07 | ModelAN | Score51.03% | Percentile76.00% | Participants26 | EvidenceB | Evaluated |
| Rank08 | ModelAC | Score48.35% | Percentile72.00% | Participants26 | EvidenceB | Evaluated |
| Rank09 | ModelMI | Score48.27% | Percentile68.00% | Participants26 | EvidenceB | Evaluated |
| Rank10 | ModelOP | Score45.36% | Percentile64.00% | Participants26 | EvidenceB | Evaluated |
| Rank11 | ModelMA | Score44.87% | Percentile60.00% | Participants26 | EvidenceB | Evaluated |
| Rank12 | ModelZA | Score44.79% | Percentile56.00% | Participants26 | EvidenceB | Evaluated |
| Rank13 | ModelGO | Score42.98% | Percentile52.00% | Participants26 | EvidenceB | Evaluated |
| Rank14 | ModelGO | Score42.55% | Percentile48.00% | Participants26 | EvidenceB | Evaluated |
| Rank15 | ModelXI | Score41.50% | Percentile44.00% | Participants26 | EvidenceB | Evaluated |
| Rank16 | ModelAC | Score40.85% | Percentile40.00% | Participants26 | EvidenceB | Evaluated |
| Rank17 | ModelOP | Score38.22% | Percentile36.00% | Participants26 | EvidenceB | Evaluated |
| Rank18 | ModelAC | Score38.22% | Percentile32.00% | Participants26 | EvidenceB | Evaluated |
| Rank19 | ModelXA | Score37.71% | Percentile28.00% | Participants26 | EvidenceB | Evaluated |
| Rank20 | ModelNV | Score37.53% | Percentile24.00% | Participants26 | EvidenceB | Evaluated |
| Rank21 | ModelXI | Score36.73% | Percentile20.00% | Participants26 | EvidenceB | Evaluated |
| Rank22 | ModelMA | Score32.10% | Percentile16.00% | Participants26 | EvidenceB | Evaluated |
| Rank23 | ModelAN | Score31.01% | Percentile12.00% | Participants26 | EvidenceB | Evaluated |
| Rank24 | ModelGO | Score29.99% | Percentile8.00% | Participants26 | EvidenceB | Evaluated |
| Rank25 | ModelXA | Score28.49% | Percentile4.00% | Participants26 | EvidenceB | Evaluated |
| Rank26 | ModelMI | Score27.89% | Percentile0.00% | Participants26 | EvidenceB | Evaluated |
The leading models and scores on this benchmark.
A closer view of the leading scores on this benchmark.
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What Finance Agent v2 measures and how its scores work.
Finance Agent v2 is a benchmark for evaluating models on financial-analysis workflows.
It measures the ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses, using the Score metric expressed as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
Common questions about Finance Agent v2.
Gemini 3.5 Flash is currently ranked first with 57.86%.
The current leaders are Gemini 3.5 Flash (57.86%), Muse Spark 1.1 (57.20%), and Claude Fable 5 (56.31%).
MiMo-V2.5 has the lowest matched official input price at $0.14 input / $0.28 output per 1M tokens.
The fastest matched records are Gemini 3.5 Flash (210.94 tok/s via Google), GPT-5.5 (134.94 tok/s via OpenAI), and Gemini 3 Flash (124.32 tok/s via Google).
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.
It measures the ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses, using the Score metric expressed as a ratio.
Yes. Higher values rank better for this benchmark.
26 model results are currently shown.
No. This benchmark is shown for reference but does not contribute to the overall score.
Ranking basisThis finance agent v2 AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
Selection summary
Gemini 3.5 Flash currently leads Finance Agent v2 with 57.86%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price, and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.