Blog
ProductIsaacus models are now up to 2.6× faster
Following an overhaul of our infrastructure and inference engine, Isaacus models are now up to 2.6× faster, with lower latencies, higher throughput, and a new status page.
Read more
ProductOur response to the US ban on Fable 5 and Mythos 5
The first US export control directive on an LLM underscores why Isaacus continues to prioritize self-hostable, sovereign AI.
Read more
EngineeringBuild a legal RAG app that won't be held in contempt
Build your first legal RAG app with Isaacus' legal embedding, reranking, and chunking models.
Read more
EngineeringHallucination-free tabular review from scratch
Learn how to build a state-of-the-art tabular review app from scratch in minutes.
Read more
ProductIntroducing AI chunking to semchunk
We're introducing a new AI chunking mode to semchunk, delivering 15% better RAG performance than its closest competitor.
Read more
ProductKanon 2 Enricher and Reranker now available in AWS Marketplace
Our flagship legal enrichment and reranking models are now publicly available in the AWS Marketplace.
Read more
ProductIntroducing Kanon 2 Reranker
We're releasing Kanon 2 Reranker, the most powerful reranking model for legal RAG to date.
Read more
ProductIntroducing Kanon 2 Enricher
We're releasing Kanon 2 Enricher, the first hierarchical graphitization model, capable of transforming unstructured data into rich, structured knowledge graphs.
Read more
