Modernbert Python Packages

laya-mlx

Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.

laya-coreml

Local Laya typed decisions on Apple Core ML and Neural Engine. Validated ports, ~5 ms short decisions on M3 Max, reproducible speed and energy benchmarks.

decision-tune

DecisionTune 1.0: a 395M decision model. Pick an option or get P(yes) with one encoder pass. PyTorch, MLX, ONNX. Apache-2.0.

laya-cuda

A lightweight CUDA inference library for Laya

tasknet

Easy modernBERT fine-tuning and multi-task learning

sokudan

Japanese System One decision model (Jev-style): typed answers and probabilities in one forward pass, no text generation. Includes bench_ja/bench_en and a Laya position-bias repro.

eeane

Run text embedding and reranking models on the Apple Neural Engine of Apple Silicon Macs. OpenAI-compatible server; ModernBERT, BERT, and XLM-RoBERTa.

julia-mlx

Julia-1 decision model on Apple silicon with MLX, at parity with the reference runtime

verbatim-rag

Provenance-first extractive RAG: return verbatim source spans with citations using local ModernBERT or optional LLM-assisted extraction.

verbatim-core

Provenance-first extractive RAG: return verbatim source spans with citations using local ModernBERT or optional LLM-assisted extraction.