Proprietary

Vals Finance Index

Updated 11/9/2026

Agentic model performance on finance work, from analyst research and financial modeling to corporate tax research, weighted by each sub-sector's share of U.S. GDP.

As of November 9, 2026, Gemini 4 Carbon ranks first on the Vals Finance Index with 67.05%, followed by Gemini 4 Argon (66.00%) and Claude Opus 5.5 (65.34%).

Vals Finance IndexGDP-weighted finance benchmark
Vals Finance Index•Nov 09, 2026
ACCURACY
Lab:
Pinch or Ctrl/Cmd + scroll to zoom. Once zoomed, scroll or swipe with two fingers to pan, or drag to pan. Shift + drag draws a box to zoom into; turn on box zoom to draw one without Shift. Plus and minus keys zoom, arrow keys pan, 0 resets.

Vals Finance Index leaderboard

Rank Model Accuracy Cost / Test Tokens / Test Duration / Test
1 Gemini 4 Carbon 67.05% $6.46 4.17m 30m38s
2 Gemini 4 Argon 66.00% $6.53 4.72m 22m41s
3 Claude Opus 5.5 65.34% $10.88 5.29m 45m00s
4 Wonder Full 64.55% $3.14 4.01m 20m06s
5 Claude Sonnet 5.5 64.53% $7.40 11.04m 43m31s
6 Claude Fable 5.2 63.58% $4.94 1.70m 8m15s
7 Qwen 3.8 Max 0902 61.66% $3.83 6.95m 1h05m
8 Inkling 2 Small 61.47% $0.28 4.37m 8m35s
9 GLM 5.3 61.41% $2.21 3.92m 35m44s
10 Wonder 61.23% $3.37 4.24m 10m12s
11 Muse Ember 60.91% $0.06 3.22m 2m36s
12 DeepSeek V4.1 Flash 1017 60.89% $0.42 8.24m 23m02s
13 Gemini 3.8 Flash 60.46% $3.70 11.05m 7m44s
14 Grok 4.7 58.83% $3.51 2.86m 20m49s
15 Hy4 58.05% $1.20 4.03m 39m29s
16 Kimi K3 57.95% $2.58 2.49m 22m24s
17 MiMo V2.6 Pro 57.85% $0.22 3.31m 30m48s
18 Hy4 Preview 57.51% $0.68 3.38m 32m15s
19 Claude Haiku 5.5 57.45% $1.39 7.47m 33m31s
20 MiMo V2.6 Flash 57.15% $0.07 3.58m 19m26s
21 DeepSeek V4.1 Flash 57.08% $0.25 3.69m 11m37s
22 Mistral Large 4 56.98% $0.56 2.07m 15m33s
23 Mistral Large 4 Preview 56.90% $1.46 3.88m 51m50s
24 Kimi K3.1 Pro 56.01% $2.51 2.27m 38m20s
25 MiniMax-M3.1-Flash 55.91% $0.25 5.19m 26m12s
26 GLM 5.3 Flash 55.83% $0.55 13.78m 33m49s
27 GPT-6.1 Sol 55.81% $1.36 927.81k 19m16s
28 GPT-6 Astra 55.64% $6.48 844.86k 16m18s
29 Step 5 Preview 55.43% $0.82 5.06m 24m42s
30 GPT-6.1 Terra 53.32% $1.55 2.53m 29m25s
31 GPT-6 Luna 53.05% $0.15 2.41m 22m01s
32 Muse Spark Open 52.99% $3.00 6.20m 41m04s
33 DeepSeek V4 Pro 0813 52.94% $0.88 3.51m 24m00s
34 MiniMax-M3 48.37% $1.22 7.52m 10m01s
35 Mercury 3 47.22% $0.10 3.38m 5m44s
36 Beam 1.0 46.50% - 3.56m 7m29s
37 Nemotron 3.5 Super VL 41.04% $0.48 8.22m 5m23s
38 Inkling 38.25% $0.97 3.19m 16m29s
39 Inkling Small 36.27% $0.18 1.55m 10m24s
Vals Finance Index Over TimeBest score from each lab, by model release date
Vals Finance Index•Nov 09, 2026
AccuracyTime
Lab:
Viewing 18 lab performance frontiers.
70%65%60%55%50%45%40%35%
Jun '26Jul '26Aug '26Sep '26Nov '26

Key Takeaways

  • Even the leaders fail much of the hardest work. The top two are within the margin of error, and both still fail many of the tax questions and a meaningful share of the financial modeling questions.
  • No model leads every component. No single model leads every component: Claude Fable 5.2 leads Finance Agent v2.1 (69.47%) and EMB v1.1 (77.38%); Gemini 4 Carbon leads Tax Agent Bench (51.78%).
  • Close behind the leader for much less. Wonder Full and DeepSeek V4.1 Flash 1017 come close to the leader at a fraction of its cost.
  • Rankings move a lot from the Vals Index. Gemini 3.8 Flash and Kimi K3 place far higher here than on the Vals Index; GPT-6 Astra and GPT-6.1 Sol place far lower.
  • Claude Sonnet 5.5 and Claude Opus 5.5 include tasks served by a fallback model after a refusal. The split icon on their rows shows the score without them.

Motivation

Finance is the largest sector of the Vals Index, but there its score is blended with coding and legal. The Finance Index looks at finance work on its own, tax included, using the same rule: each benchmark maps to an industry sub-sector, and each sub-sector is weighted by its share of U.S. GDP.


Methodology

Components

  • Securities & Investments (NAICS 523): Finance Agent v2.1, analyst research on public company filings, and the Excel Modeling Benchmark v1.1, building and editing financial models in Excel (LBO, DCF, M&A, comparables, operating models).
  • Tax (NAICS 5412): Tax Agent Bench, US corporate tax research questions answered with cited authority.

Each component column is that benchmark’s published standalone score. Tax Agent Bench contributes all-pass Accuracy, as it does on the Vals Index. A model is scored only when all three components have a result for it.

Weights

The Bureau of Economic Analysis splits finance and insurance into sub-sectors and files accounting and tax preparation under professional services. Each is weighted by its value added as a share of U.S. GDP, from BEA’s GDP by Industry accounts (annual 2025, September 2026 update; the finance and insurance total is also on FRED).

Sub-sector (NAICS)Share of GDPComponents
Securities, commodity contracts, and investments (523)1.6%Finance Agent v2.1, EMB v1.1
Accounting, tax preparation, bookkeeping, and payroll (5412)0.5%Tax Agent Bench

Formula

Each sub-sector is the average of its benchmarks, and the sub-sectors are combined by weight:

Securities = AVG(FinanceAgentV2_1, EMBV1_1)
Tax = TaxAgentBench
Finance_Index = (1.6 * Securities + 0.5 * Tax) / 2.1

Updates

10/21/2026

Released Vals Finance Index v1. Finance Agent v2.1 and EMB v1.1 form the Securities & Investments sub-sector (weight 1.6); Tax Agent Bench forms the Tax sub-sector (weight 0.5).