Isaacus

Team

company

Verified

https://isaacus.com/

AI & ML interests

We build legal AI models for legal tech firms that help them ship smarter products, faster.

Papers

Legal RAG Bench: an end-to-end benchmark for legal RAG

The Massive Legal Embedding Benchmark (MLEB)

View all Papers

Articles

Introducing Kanon 2 Enricher — the world’s first hierarchical graphitization model

Mar 3

• 7

Introducing Legal RAG Bench

Feb 20

• 14

Australian-made LLM beats OpenAI and Google at legal retrieval

Oct 23, 2025

• 28

Introducing the Massive Legal Embedding Benchmark (MLEB)

Oct 17, 2025

• 24

View all articles

isaacus 's models 7

isaacus/kanon-2-tokenizer

Updated Feb 16 • 4

isaacus/kanon-tokenizer

Updated Oct 15, 2025 • 5

isaacus/emubert

Fill-Mask • 0.1B • Updated Jun 12, 2024 • 1.8k • • 10

isaacus/open-australian-legal-llm

Text Generation • 2B • Updated Dec 15, 2023 • 137 • 13

isaacus/open-australian-legal-phi-1_5

Text Generation • Updated Dec 15, 2023 • 11 • 2

isaacus/open-australian-legal-distilgpt2

Text Generation • Updated Dec 15, 2023 • 13 • 2

isaacus/open-australian-legal-gpt2

Text Generation • 0.1B • Updated Dec 15, 2023 • 246 • 5

AI & ML interests

Papers

Articles

Announcing the Blackstone Graph

Build a legal RAG app that won't be held in contempt

Building Harvey-style tabular review from scratch, but better

Introducing AI chunking to semchunk

Kanon 2 Reranker: the most powerful reranker for legal RAG

Introducing Kanon 2 Enricher — the world’s first hierarchical graphitization model

Introducing Legal RAG Bench

Australian-made LLM beats OpenAI and Google at legal retrieval

Introducing the Massive Legal Embedding Benchmark (MLEB)

Team members 4

isaacus 's models 7 Sort: Recently updated

isaacus 's models 7