Skip to content

Groot 1 · v0.1.0

Choose a tag to compare

@leofmarciano leofmarciano released this 22 Oct 13:30
· 2 commits to main since this release

First public release of Groot 1, an open typed-decision model by Marciano Theory SI.

Highlights

  • Typed decisions (2,000 questions): Groot 1 S 75.2% · Jev 73.7% · Julia 1 72.7%
  • Calibration: typed-decision Brier 0.072, against 0.156 for Jev and 0.336 for Julia 1
  • Long context: 96–100% needle accuracy from 1k to 64k tokens
  • System One compatible server (/v1/systemone), up to 255 options per call
  • Pure Bend engine matching PyTorch to within 1e-7

What's inside

  • groot1/: model, serializer with BM25 state retrieval, trainer, evaluation, server, export
  • bend/: the full forward pass in Bend (JS host and native binary)
  • scripts/benchmark.py: three-way benchmark harness; raw results in metrics/

Known limits

Classification results are in-domain. Groot trails Jev on BoolQ, ASSIN2 and HateBR, and is slower than Jev's API above 16k tokens. Weights will be published separately. See the README for the full protocol.

🤖 Generated with Claude Code

https://claude.ai/code/session_01SeDoQ4GNRjX8TpKtPhNnMr