Proprietary

Vals Index

Updated 10/13/2026

A single measure of AI's potential economic impact — agentic model performance across finance, coding, legal, and tax tasks, weighted by each sector's share of U.S. GDP.

As of October 13, 2026, Gemini 4 Argon ranks first on the Vals Index with 68.90%, followed by Claude Sonnet 5.5 (67.04%) and Claude Opus 5.5 (66.97%).

Vals IndexGDP-weighted benchmark
ACCURACY

Vals Index leaderboard

Rank Model Accuracy Cost / Test Input / Output Cost Input / Output Tokens Duration
1 Gemini 4 Argon 68.90% $15.68 $4 / $20 13.40B / 142.01M 46m33s
2 Claude Sonnet 5.5 67.04% $21.34 $2 / $10 40.38B / 598.68M 1h18m
3 Claude Opus 5.5 66.97% $32.14 $4 / $20 28.57B / 477.68M 1h19m
4 Claude Fable 5.1 65.83% $28.71 $10 / $50 18.94B / 202.22M 1h17m
5 Wonder Full 64.57% $8.56 $4 / $20 15.29B / 102.21M 56m36s
6 Claude Opus 5 63.67% $19.31 $5 / $25 14.71B / 146.95M 57m42s
7 GPT-6 Astra 63.13% $18.46 $10 / $50 3.93B / 94.52M 27m48s
8 Claude Fable 5 61.39% $29.59 $10 / $50 10.43B / 124.29M 42m24s
9 GPT-6.1 Sol 61.15% $3.24 $2 / $10 4.50B / 104.77M 43m10s
10 Wonder 61.06% $11.43 $4 / $20 16.64B / 96.28M 37m08s
11 Muse Spark 1.3 Max 58.16% $3.79 $1.25 / $4.25 11.28B / 94.12M 23m34s
12 GPT-5.6 Sol 58.01% $14.24 $4 / $20 6.46B / 145.31M 37m23s
13 GPT-6 Sol 57.54% $7.58 $2 / $10 8.97B / 104.59M 29m11s
14 MiMo V2.6 Pro 55.20% $0.41 $0.435 / $0.87 8.35B / 117.82M 1h07m
15 Claude Opus 4.8 55.10% $13.14 $5 / $25 7.98B / 118.22M 39m48s
16 Grok 4.7 54.95% $12.11 $2 / $6 7.28B / 113.97M 35m30s
17 Gemini 3.8 Flash 54.89% $5.71 $1.5 / $7.5 12.99B / 114.57M 51m52s
18 Claude Haiku 5.5 54.44% $3.00 $0.1 / $0.5 21.99B / 748.63M 47m24s
19 GLM 5.3 53.51% $7.25 $1.4 / $4.4 12.09B / 211.64M 1h13m
20 MiMo V2.6 Flash 53.23% $0.20 $0.14 / $0.28 11.54B / 124.57M 58m02s
21 Muse Spark 1.3 53.20% $3.38 $1.25 / $4.25 8.12B / 77.28M 32m46s
22 GPT-5.6 Terra 53.09% $5.78 $2 / $12 7.64B / 231.74M 35m39s
23 Grok 4.6 52.09% $4.49 $2 / $6 4.42B / 103.25M 37m01s
24 Claude Sonnet 5 51.90% $13.69 $2 / $10 26.59B / 166.56M 54m15s
25 GPT-5.6 Luna 51.69% $0.82 $0.2 / $1.2 9.48B / 226.96M 29m27s
26 DeepSeek V4.1 Flash 51.32% $0.33 $0.3 / $1.2 10.05B / 138.82M 27m53s
27 Gemini 3.7 Flash 51.27% $4.67 $1.5 / $7.5 9.15B / 89.08M 47m27s
28 GPT-6 Luna 51.22% $0.43 $0.1 / $0.5 9.74B / 201.05M 30m34s
29 Ember-1 50.94% $9.66 $3 / $15 13.00B / 96.37M 1h10m
30 Kimi K3 50.30% $6.44 $3 / $15 7.91B / 86.25M 1h08m
31 Hy4 Preview 49.94% $1.44 $0.834 / $2.501 10.71B / 179.91M 1h01m
32 Step 5 Preview 49.35% $2.63 $1 / $2.7 16.89B / 132.96M 1h10m
33 Muse Spark 1.2 49.29% $1.94 $1.25 / $4.25 5.06B / 63.06M 22m56s
34 Qwen 3.8 Max 48.27% $4.96 $2 / $6 7.75B / 148.01M 1h41m
35 Mistral Large 4 48.11% $13.78 $1.36 / $4.18 10.46B / 161.01M 1h42m
36 DeepSeek V4 Flash 0731 47.96% $0.85 $0.44 / $1.32 14.50B / 131.10M 43m06s
37 DeepSeek V4 Pro 0813 47.63% $3.35 $1.32 / $3.96 18.05B / 152.98M 1h04m
38 Gemini 3.5 Flash 44.79% $3.30 $1.5 / $9 6.90B / 86.21M 18m17s
39 Grok 4.5 44.72% $2.42 $2 / $6 2.26B / 55.50M 15m43s
40 GLM 5.3 Flash 44.58% $1.03 $0.075 / $0.25 30.91B / 187.17M 1h11m
41 DeepSeek V4 38.63% $1.31 $1.32 / $3.96 8.17B / 94.39M 31m53s
42 MiniMax-M3 36.53% $3.08 $0.6 / $2.4 9.64B / 70.31M 38m55s
43 MiMo V2.5 Pro 33.84% $0.16 $0.435 / $0.87 3.11B / 46.63M 34m33s
44 Gemini 3.1 Pro Preview (02/26) 33.44% $2.18 $2 / $12 3.12B / 27.31M 9m52s
45 GPT 5.4 Mini 33.17% $1.45 $0.75 / $4.5 5.52B / 219.60M 29m47s
46 Inkling 28.67% $1.46 $1 / $4.05 4.51B / 69.35M 20m07s
47 Inkling Small 25.46% $0.28 $0.3 / $1.2 2.53B / 67.04M 21m02s
48 Mercury 2.5 8.86% $0.29 $0.2 / $0.75 1.94B / 23.07M 5m44s

Motivation

As AI capabilities rapidly advance, understanding their potential to transform economic sectors has become critical for organizations making deployment decisions. Unlike existing aggregated metrics that treat all capabilities equally, the Vals Index is designed to reflect the potential impact of AI models on the U.S. economy. We accomplish this by computing a weighted average of model performance across key sectors, where each sector’s weight is proportional to its share of U.S. GDP.

Vals AI has developed a comprehensive suite of benchmarks measuring AI models’ ability to perform real-world tasks across finance, coding, legal, and tax work — benchmarks that no other party has full access to. The Vals Index aggregates six private and two public benchmarks to provide high signal into the real-world tradeoffs between capability, latency, and cost of deploying AI systems.

Results

Industry Average Accuracy Comparison

Key Takeaways

AI models are advancing rapidly in their ability to handle complex, real-world tasks across critical economic sectors. The results demonstrate that frontier models are becoming increasingly capable at automating work in finance, coding, legal, and tax services — domains that collectively represent a substantial portion of economic activity.

Gemini 4 Argon leads at 68.90%, ahead of Claude Sonnet 5.5 (67.04%) and Claude Opus 5.5 (66.97%). No single model leads every component: Claude Fable 5.1 leads EMB (76.67%) and Tax Agent (77.64%); Claude Sonnet 5.5 leads Vibe Code Bench (92.39%) and Code Migration (70.13%); Gemini 4 Argon leads Finance Agent v2 (65.40%); Claude Opus 5.5 leads Terminal-Bench 4.0 (65.15%); Muse Spark 1.2 leads HLAB (25.42%); Muse Spark 1.3 Max, Claude Opus 5, and Claude Fable 5.1 tie on Legal Research (55.29%).

Scores for Sonnet 5.5, Opus 5.5 and Fable 5.1 include some component tasks served by a fallback model after a provider refusal; counting those as failures lowers them to 65.85%, 65.05% and 64.59% respectively, with the order unchanged. Gemini 4 Argon had no fallbacks. The split icon on each affected row itemizes them.

Methodology

Benchmark Selection, Economic Weighting, and Formula

The Vals Index aggregates performance across four sectors — Finance, Coding, Legal, and Tax — each weighted in proportion to its share of U.S. GDP, using Bureau of Economic Analysis value-added-by-industry data (as published on FRED). While this represents a vast oversimplification of how AI might impact the economy, it provides a useful proxy for measuring the potential economic significance of model capabilities:

Finance: Finance & Insurance, ~8.0% of U.S. GDP

Coding: Information sector (the closest GDP proxy for software work), ~5.6% of U.S. GDP

Legal: Legal services, ~1.2% of U.S. GDP

  • Legal Research Bench: Case and statute research with citation-backed answers
  • HLAB: Harvey’s Legal Agent Benchmark, long-horizon legal work product creation

Tax: Accounting and tax preparation services (NAICS 5412), ~0.5% of U.S. GDP

  • Tax Agent Bench: Research-grade US corporate tax questions answered with cited authority

Each sector is the average of its benchmarks, and the sectors are combined with the following formula:

Finance = AVG(FinanceAgentV2, EMB)
Coding = AVG(TBench4, VibeCodeBench, CodeMigration)
Legal = AVG(LegalResearch, HLAB)
Tax = TaxAgentBench
Vals_Index = (8.0 * Finance + 5.6 * Coding + 1.2 * Legal + 0.5 * Tax) / 15.3

Component Scores

Each benchmark column on the index is that benchmark’s own published standalone score — the same number, under the same methodology. The exception is Code Migration: there, the index scores a fixed subset of the published run: 50 of the 120 CLI migration tasks, plus all 10 COBOL tasks, weighted 75% CLI and 25% COBOL to match the standalone benchmark’s balance. This subset was chosen by Monte Carlo search, scored against the full 120-task leaderboard. The chosen subset reproduces the full 120-task result to Spearman 0.99 and 0.8 points of mean absolute accuracy error, with no model moving more than three ranks; on models held out of the selection, mean absolute error is 1.1 points.

Updates

9/25/2026

Released Vals Index v2.1:

  • Added a Tax sector. Tax Agent Bench, a private benchmark of research-grade US corporate tax questions, enters as its own sector weighted 0.5 — the GDP share of accounting and tax preparation services (NAICS 5412), which the BEA files under professional services rather than Finance & Insurance. The denominator moves from 14.8 → 15.3.
  • Updated the coding bucket to Terminal-Bench 4.0. Terminal-Bench 4.0 replaces 2.1 in the Coding sector; Finance and Legal components are unchanged. Terminal-Bench 4.0 is substantially harder, so index scores are lower than under v2.

8/13/2026

Released Vals Index v2:

  • Added EMB in place of CorpFin. A private benchmark for building complex financial models in Excel.
  • Added Code Migration to the coding bucket. A private benchmark involving porting projects to a set of four languages, including COBOL modernization.
  • Added Legal Research Bench to the returning Legal sector. A private benchmark involving answering legal questions grounded in case law.
  • Added HLAB to the returning Legal sector. Harvey’s Legal Agent Benchmark, long-horizon legal work product creation.
  • Removed SWE-Bench Verified from the coding bucket. SWE-Bench has become saturated. Coding now averages Terminal-Bench 2.1, Vibe Code Bench, and Code Migration equally.

5/27/2026

  • Updated the coding bucket to Terminal-Bench 2.1. The Vals Index now evaluates coding performance with Terminal-Bench 2.1 while keeping the same 0.25 coding weight.

5/13/2026

  • Swapped Finance Agent to Finance Agent v2. Finance now uses the Finance Agent v2 index subset, averaging three runs per model.

5/4/2026

  • Added Vibe Code Bench to the coding bucket. Coding is now a weighted average of three benchmarks (SWE-Bench Verified 0.25, Terminal-Bench 2.0 0.25, Vibe Code Bench 0.5), giving end-to-end app-building tasks half of the coding signal. VCB is evaluated on a 22-task subset selected for coverage across UI, data, and workflow patterns.
  • Removed the Law sector (CaseLaw). The CaseLaw benchmark had become saturated, and was no longer providing useful differentiation between models. Consequently, it was removed from the index. The denominator was rebalanced from 3.7 → 3.4 to reflect the dropped 0.3 law weight, and the Industry Average chart no longer displays a Law column.