Hey all, We're launching Kanon 2 Embedder, a state-of-the-art legal domain embedding model pretrained and finetuned on millions of laws, cases and contracts from around the world. Kanon 2 Embedder ranks first on the Massive Legal Embedding Benchmark (MLEB) as of 19 October 2025. On MLEB, it is 9% more accurate than OpenAI Text Embedding 3 Large and 6% more accurate than Google Gemini Embedding, all while running 30% faster than both LLMs. Kanon 2 Embedder leads a field of 20 LLMs, including Qwen3 Embedding 8B, IBM Granite Embedding R2, and Microsoft E5 Large Instruct. Importantly, unlike other high-scoring models on MLEB, Kanon 2 Embedder is also privacy and security friendly — none of your data is used to train our models by default — and can be self-hosted for enterprises with heightened security or reliability requirements.

Kanon 2 Embedder by Isaacus

A new state-of-the-art legal embedding model

A new state-of-the-art legal embedding model