Benchmarks · Artificial Analysis
Artificial Analysis Intelligence Index
Artificial Analysis composite index of general intelligence (reasoning, knowledge, coding, agentic work). Not legal. PrudentScore v1.1 uses version 4.3.2; Artificial Analysis moved to v5.0 on 27 October 2026 (different scale).
- Models measured
- 23
- Version
- 4.3.2
- Measured
- 20 October 2026
01
In running text
Artificial Analysis measured 23 models from the register on version 4.3.2, 14 of which pass our compliance gate. Of that group, Claude Opus 5.5 ranks highest at 57.6, position 1 of 23. Source: Artificial Analysis (artificialanalysis.ai/models), accessed 1 October 2026 [37].
Figures from this source appear in running text only, with attribution, as that party’s terms require [9]. There is no table of them.
Other benchmarks
- Legal Research Bench
- Harvey LAB (Vals implementation)
- LegalBench (Vals leaderboard)
- CaseLaw v2
- TaxEval v2
- Tax Agent Bench
- CorpFin v2
- Immigration Bench
- Access to Justice
- Public Benefits Bench
- Vals Legal Index
- Vals Index
- GPQA Diamond (Vals-run)
- MMLU Pro (Vals-run)
Claude and Anthropic, GPT and OpenAI, Gemini and Google, Grok and xAI, Kimi and Moonshot AI, Muse and Meta, Mistral, DeepSeek, GLM and Z.ai, Harvey, Vals AI and Artificial Analysis are trademarks of their respective owners. They are named for identification only and this implies no partnership, sponsorship or endorsement.