Skip to main content
Latest Update
2026-11-19 10:30 UTC

Muse Wonder tops Legora BAR benchmark in November Analysis

Muse Wonder tops Legora BAR benchmark in November Analysis

According to Legora, Muse Wonder earned the highest overall quality score in the November BAR benchmark for legal tasks, leading in analysis, extraction, drafting, and editing.

Source

Analysis

Legora released its November BAR benchmark results evaluating AI models on real-world legal tasks, with Muse Wonder earning the highest overall quality score and leading in analysis and answering, extraction, drafting, and editing according to the announcement from Legora shared via AI at Meta.

GPT-6.2 Astra led comparison and review tasks with overall quality 4% above the November model average and cost per case 20% below average, while GPT-6.2 Sol delivered above-average quality at 78% below average cost per case. These outcomes highlight AI tools that can enhance developer and legal professional productivity by handling complex business tasks such as document analysis and drafting, reshaping workflows in the legal sector and signaling competitive differentiation among frontier models in the broader AI ecosystem.

AI at Meta

@AIatMeta

Together with the AI community, we are pushing the boundaries of what’s possible through open science to create a more connected world.