Artificial Analysis Launches Industry Indices to Benchmark AI on Professional Tasks

Artificial AnalysisArtificial Analysis

Artificial Analysis released six new Capability Indices evaluating AI models across Finance, Legal, Healthcare, Strategy, Engineering, and Economics. The benchmarks use occupational data to weight model performance based on the actual frequency of professional tasks like contract review and clinical documentation. Results reveal a massive frontier premium, with top-tier models costing over 100x more than mid-tier alternatives for incremental accuracy gains.

Artificial Analysis, an independent AI benchmarking company, launched six vertical-specific Capability Indices to evaluate model performance in professional domains. These indices cover Finance, Legal, Healthcare, and others. They join existing metrics like the Artificial Analysis Agentic Index to provide a granular view of model utility in professional knowledge work.
Leading model
Claude Fable 5
Open-weights leader
GLM-5.2 (max)
Cost variance
Over 100x per task
Latency variance
~15x within indices
Industry domains
Finance, Legal, Healthcare, and more

The methodology grounds evaluations in O*NET occupational data, weighting benchmarks by how often specific capabilities appear in real-world jobs. This reveals a steep "frontier premium" where models like Claude Fable 5 lead every index but can cost over 100x more per task than mid-tier models for a 12-point scoring advantage.

You can use these indices to optimize model selection based on cost and accuracy for specific industries. Results show that models like GLM-5.2 (max) currently lead open-weights performance in most categories. Full methodology and benchmarks are available on the Artificial Analysis website, with metrics updated as new models launch.

View the full update on artificialanalysis.ai
Artificial Analysis
Artificial Analysis
@ArtificialAnlys
X

Introducing six new Artificial Analysis Capability Indices for comparing model capabilities across key industry domains The new industry indices cover Finance & Accounting, Legal, Healthcare & Medical, Strategy & Ops, Engineering, and Economics. We aim to capture the common capabilities required across knowledge work domains and evaluate how well current models meet those needs. Each index is grounded in common tasks from O*NET occupational classifications. Tasks range from financial modeling, to legal research and contract review, to clinical decision support and patient documentation. We derive capabilities from each task, select the benchmarks that best represent the work, and weight by how often each capability appears across the domain. This means rethinking the Artificial Analysis benchmark suite for each domain and slicing evaluations to relevant domain tasks. Every component benchmark is run independently by Artificial Analysis. The industry indices join the existing skill-based Agentic and Coding indices, which measure capabilities that cut across every domain. Key Results ➤ Leading models: Claude Fable 5 (with Opus 4.8 fallback) leads all eight indices, with Claude Opus 4.8 (max) in second on six of eight Capability Indices and GPT-5.5 (xhigh) on two. Below the top two, rankings reshuffle substantially by domain between Gemini 3.5 Flash, Gemini 3.1 Pro Preview, GPT-5.5 (xhigh), Claude Sonnet 5 (max), and GLM-5.2 (max). ➤ Open weights leading models: Among open weights models, GLM-5.2 (max) leads on five of the six industry indices, ranking as high as fifth overall on the Artificial Analysis Engineering Index (53), within 2 points of Claude Sonnet 5 (max, 55) and GPT-5.5 (xhigh, 55). DeepSeek V4 Pro (max, 38) takes the open weights lead on Artificial Analysis Strategy & Ops Index. ➤ Cost efficiency: DeepSeek V4 Flash (max) completes tasks for <$0.04 across all six indices while scoring mid-pack, and GLM-5.2 (max) leads open weights score with a Cost per Task of $0.26 to $0.58. Frontier capability comes at a steep premium: on the Artificial Analysis Strategy & Ops Index, Claude Fable 5 (with Opus 4.8 fallback, $3.48) scores 12 points above DeepSeek V4 Pro (max, $0.03) at over 100x the Cost per Task. ➤ Time per Task: Time per Task spreads roughly 15x within each index, from 1.1 minutes for Nova 2.0 Pro Preview (medium) to 16.7 minutes for Claude Sonnet 5 (max). Speed shows a similar frontier to cost: on the Artificial Analysis Legal Index, Gemini 3.1 Pro Preview (0.8 minutes) completes tasks ~7x faster than Claude Fable 5 (with Opus 4.8 fallback, 5.4 minutes), while scoring within 11 points.

16retweets237likes
View on X

Still wondering? A few quick answers below.

The new Capability Indices are six industry-specific benchmarks designed to evaluate AI models on professional tasks. They cover Finance and Accounting, Legal, Healthcare and Medical, Strategy and Operations, Engineering, and Economics. These indices measure how well models handle domain-specific work like financial modeling, contract review, and clinical documentation.

Each index is grounded in common tasks from O*NET occupational classifications. Artificial Analysis derives specific capabilities from these tasks, selects relevant benchmarks to represent the work, and weights them by how often each capability appears in that domain. Every component benchmark is run independently by the company to ensure objective results.

Claude Fable 5 currently leads all eight indices, including the new industry-specific ones and existing skill-based benchmarks. Claude Opus 4.8 and GPT-5.5 follow closely in the top rankings. Among open-weights models, GLM-5.2 leads in five of the six industry indices, showing performance competitive with proprietary frontier models in engineering tasks.

Frontier capabilities come at a significant premium, often costing over 100 times more per task than mid-tier alternatives. For example, on the Strategy and Operations Index, Claude Fable 5 costs roughly 3.48 dollars per task, while DeepSeek V4 Pro costs only 0.03 dollars per task despite scoring within 12 points of the leader.

Time per task varies by roughly 15 times within each index. On the Legal Index, Gemini 3.1 Pro Preview completes tasks in 0.8 minutes, which is about seven times faster than Claude Fable 5. This data helps users choose models based on whether they prioritize absolute accuracy or faster, more cost-effective execution.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Artificial Analysis

See all AI news & updates from Artificial Analysis →

Share this update