Lexiel 99.3%: How We Beat Maite AI and ChatGPT on the Spanish Legal Benchmark

The Problem We Cannot Ignore: Legal Hallucinations

In 2023, a New York attorney filed a brief stuffed with ChatGPT-invented court decisions. The judge imposed a $5,000 fine and the case went viral. Spanish bar associations issued warnings. The question hung in the air: is it safe to use AI for real legal work?

The answer isn't simply yes or no. It depends entirely on which AI you use and how it's built.

The Test: 150 Questions from the Bar Access Exam

To answer that question objectively, we put four tools through Spain's official bar access exam (2023, 2024 and 2025 editions, 150 multiple-choice questions). We chose this exam for three reasons:

1. Maximum difficulty: It's the reference legal exam in Spain, covering all branches of the legal system.
2. Verifiable answers: Every correct answer is published by the Ministry of Justice with a definitive answer key.
3. No shortcut possible: Questions require genuine legal reasoning about current law, not fact memorization.

Tools tested: Lexiel, Claude (without legal RAG), Gemini 2.5 Flash (without RAG), and ChatGPT-4o.

The Results

Evaluation was conducted using a blind methodology against the definitive answer keys published by the Ministry of Justice (2023, 2024 and 2025 editions).

Why Lexiel Beats General-Purpose Models

The difference isn't the underlying AI model : Lexiel uses state-of-the-art AI models. The difference is the verified legal corpus powering it via RAG (Retrieval-Augmented Generation):

• 69 BOE statutes: Civil Code, LEC, LECr, Criminal Code, LOPJ, Constitution, regional statutes and more
• 1,868 rulings (Supreme Court + Constitutional Court) indexed with semantic embeddings tuned to Spanish legal terminology
• 1,176 Constitutional Court rulings from 1980 through 2025
• ~95,000 semantic fragments available for real-time retrieval

When Lexiel answers about Article 609 of the Civil Code, it doesn't generate freely: it retrieves the exact fragments from the real statute and builds the answer on that verified foundation. If it can't verify something in the corpus, it says so explicitly.

Why Lexiel Is Different from Maite AI

Maite AI is a legal research assistant. Lexiel goes beyond with:

1. Verified corpus of 1,950 sources: 69 BOE statutes, 1,868 rulings (Supreme + Constitutional Court), CENDOJ case law, all indexed with semantic embeddings.
2. Hybrid RAG: Vector + BM25 search with Reciprocal Rank Fusion (RRF), per-article deduplication, and verification against official sources.
3. Complete law firm platform: Legal CRM, Verifactu invoicing, client portal, procedure engine for automated briefs, time tracking, and automatic deadline management.
4. Multi-jurisdiction: Legal corpus spanning 20 countries (Spain, Mexico, Colombia, Argentina, Chile, Peru, Uruguay, Ecuador, Panama, Costa Rica, Venezuela, Bolivia, Guatemala, Honduras, El Salvador, Paraguay, Nicaragua, Brazil, Dominican Republic, EU).

What the Benchmark Does NOT Measure

Honesty matters. This test measures pure legal accuracy. It doesn't measure:

• Response speed (Lexiel uses real-time streaming)
• Integration with daily law firm workflows
• Usability and learning curve
• Productivity features beyond chat

In all those dimensions Lexiel also leads, but those are aspects every attorney should evaluate with their own 14-day free trial.

Conclusion

The benchmark confirms what our users already know from daily work: Lexiel doesn't invent. It cites. And when it can't verify, it says so.

If you use legal AI for research, brief drafting, or client consultations, it's worth demanding the same standard you'd demand from a colleague: cite the source, and make sure the source is real.

View full benchmark methodology →