# Vals Legal Index v1 (Vals Legal Index)

> Vals AI composite of agentic legal work: legal research, drafting and reviewing documents, and tax law, with practice areas weighted by their share of U.S. law-office revenue.

Canonical page: https://benchlm.ai/benchmarks/vals-legal-index

- Category: [Agentic](/agentic)
- Last updated: October 27, 2026

Evaluation results published by [Vals AI](https://www.vals.ai/benchmarks/legal_index). Published leaderboard aggregates and evaluation settings only. Dataset contents are not included.

## About Vals Legal Index

- Year: 2026
- Tasks: 3 legal benchmarks across 8 practice areas plus tax
- Format: Revenue-weighted accuracy
- Difficulty: Agentic legal research, document drafting and review, and tax research
- Paper: [Vals Legal Index leaderboard and methodology](https://www.vals.ai/benchmarks/legal_index)

Vals published v1 on October 24, 2026. Legal Research Bench (all-pass accuracy by area of law) and HLAB-Vals-1.1 (criteria pass rate by practice area) score eight practice areas, averaged where both cover an area; the areas combine by their share of 2022 law-office revenue from the Census Bureau's Economic Census. Tax Agent Bench carries the taxation sub-line: Legal_Index = 0.978 x practice areas + 0.022 x Tax Agent Bench. We mirror the published aggregates as a display-only composite, outside overall and category rankings.

Vals Legal Index is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (30 models)

| Rank | Model | Configuration | Creator | Score |
|------|-------|---------------|---------|-------|
| 1 | [Muse Wonder Full](/models/muse-wonder-full) | max reasoning | Meta | 77.86% |
| 2 | [Gemini 4 Argon](/models/gemini-4-argon) | high reasoning | Google | 76.72% |
| 3 | [Gemini 4 Carbon](/models/gemini-4-carbon) | high reasoning | Google | 75.83% |
| 4 | [Muse Wonder](/models/muse-wonder) | max reasoning | Meta | 75.26% |
| 5 | [GLM-5.3](/models/glm-5-3) | max reasoning | Z.AI | 74.58% |
| 6 | [Qwen3.8 Max Internal](https://www.vals.ai/models/alibaba_qwen3.8-max-internal) | — | Alibaba | 73.90% |
| 7 | [Grok 4.7](/models/grok-4-7) | xhigh reasoning | xAI | 72.60% |
| 8 | [Claude Opus 5.5](/models/claude-opus-5-5) | — | Anthropic | 72.33% |
| 9 | [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) | — | Xiaomi | 72.14% |
| 10 | [DeepSeek V4.1 Flash 1017](/models/deepseek-v4-1-flash) | max reasoning | DeepSeek | 71.86% |
| 11 | [Kimi K3](/models/kimi-k3) | max reasoning | Moonshot AI | 71.72% |
| 12 | [Hy4 preview](/models/hy4-preview) | — | Tencent | 71.70% |
| 13 | [GLM-5.3-Flash](/models/glm-5-3-flash) | max reasoning | Z.AI | 71.05% |
| 14 | [Claude Haiku 5.5](/models/claude-haiku-5-5) | — | Anthropic | 70.29% |
| 15 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | high reasoning | Google | 69.78% |
| 16 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | max reasoning | DeepSeek | 69.01% |
| 17 | [MiMo-V2.6-Flash](/models/mimo-v2-6-flash) | — | Xiaomi | 68.99% |
| 18 | [GPT-6.1 Sol](/models/gpt-6-1-sol) | max reasoning | OpenAI | 68.72% |
| 19 | [Muse Spark Open](/models/muse-spark-open) | xhigh reasoning | Meta | 68.51% |
| 20 | [GPT-6 Astra](/models/gpt-6-astra) | max reasoning | OpenAI | 68.39% |
| 21 | [Claude Sonnet 5.5](/models/claude-sonnet-5-5) | — | Anthropic | 68.02% |
| 22 | [MiniMax M3.1 Flash](/models/minimax-m3-1-flash) | — | MiniMax | 67.07% |
| 23 | [Kimi K3.1 Pro](/models/kimi-3-1-pro) | max reasoning | Moonshot AI | 67.04% |
| 24 | [Mistral Large 4](/models/mistral-large-4) | high reasoning | Mistral | 65.61% |
| 25 | [Mercury 3](/models/mercury-3) | high reasoning | Inception | 62.96% |
| 26 | [GPT-6 Luna](/models/gpt-6-luna) | max reasoning | OpenAI | 62.20% |
| 27 | [Beam 1.0](https://www.vals.ai/models/reflection_Beam-1.0) | — | Reflection | 61.80% |
| 28 | [GPT-6.1 Terra](/models/gpt-6-1-terra) | max reasoning | OpenAI | 58.64% |
| 29 | [Inkling-Small](/models/inkling-small) | 0.99 reasoning | Thinking Machines Lab | 57.83% |
| 30 | [Inkling](/models/inkling) | 0.99 reasoning | Thinking Machines Lab | 54.16% |

## FAQ

### What does Vals Legal Index measure?

Vals AI composite of agentic legal work: legal research, drafting and reviewing documents, and tax law, with practice areas weighted by their share of U.S. law-office revenue.

### Which model leads the published Vals Legal Index snapshot?

Muse Wonder Full currently leads the published Vals Legal Index snapshot with a score of 77.86%.

### How many models are evaluated on Vals Legal Index?

The October 27, 2026 contains 30 AI models.

### Does Vals Legal Index affect BenchLM's overall score?

Not directly. Vals Legal Index is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.