HRM-Text is a 1B text generation model based on the HRM architecture, strengthened by task completion and latent space reasoning.
-
Updated
Sep 21, 2026 - Python
HRM-Text is a 1B text generation model based on the HRM architecture, strengthened by task completion and latent space reasoning.
Implementation of the Hierarchical Reasoning Model (HRM), applied to a pathfinding task, plus performance study.
Official PyTorch implementation of "Latent Reasoning in TRMs is Secretly a Policy Improvement Operator" (ICML 2026)
HRM-sMoE LLM training toolkit.
Just testing if HRM Generalizes better compared to others in Vision Task
A laptop version of Hierarchical Reasoning Model
Implementation of the Hierarchical Reasoning Model (HRM), based on the research paper “Hierarchical Reasoning Model” (2025) by Guan Wang, Jin Li, Yuhao Sun, Xing Chen, Changling Liu, Yue Wu, Meng Lu, Sen Song, and Yasin Abbasi Yadkori (arXiv:2506.21734).
HRM models made easy.
HRM-Text is a 1B text generation model based on the HRM architecture, strengthened by task completion and latent space reasoning.
Mathematical resolution to the Sapient AI HRM scaling wall (Issue #80). Open-sourcing Poincaré Deep Equilibrium (P-DEQ) and Grid-Cell Topological Anchoring to solve ZOH/Dirac Delta gradient shatter.
Hierarchical Reasoning Model (TinyHrm) in PyTorch with a toolkit: one-step gradient, deep supervision, maze and Sudoku tasks
To associate your repository with the hierarchical-reasoning-model topic, visit your repo's landing page and select "manage topics."