# sapientinc/HRM

Hierarchical Reasoning Model Official Release

Repository: https://github.com/sapientinc/HRM
Canonical: https://ross.abutalabs.com/products/hrm
Homepage: https://sapient.inc
Language: Python
License: Apache-2.0
License Family: permissive
Topics: brain-inspired-ai, deep-learning, large-language-models, reasoning
Last push: 2026-03-31T23:08:31+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 75, release rhythm 35, longevity 30
- inputs: {"age_days": 421, "days_push": 155, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12619, forks 1825 (observed 2026-08-28T04:10:59.429674+00:00)

## What it is
Official PyTorch implementation of the Hierarchical Reasoning Model (HRM), a 27M-parameter recurrent architecture with high-level and low-level modules that performs sequential reasoning in a single forward pass without chain-of-thought data. It achieves strong results on Sudoku, maze pathfinding, and the ARC benchmark with only ~1000 training samples and no pre-training.

## Use cases
- reproduce HRM results on Sudoku and maze reasoning tasks
- train a small reasoning model on the ARC benchmark
- experiment with brain-inspired hierarchical recurrent architectures
- research latent-space reasoning without chain-of-thought supervision
- benchmark small models against large LLMs on reasoning tasks

## When to choose
- you need a compact reasoning model trainable on tiny datasets without pre-training
- you are researching recurrent or hierarchical architectures for reasoning
- you have CUDA GPUs and want to reproduce the paper's ARC/Sudoku/maze results

## When to avoid
- you need a production text-generation or chat model out of the box
- you lack NVIDIA GPUs or cannot build CUDA/FlashAttention extensions
- you want a plug-and-play library rather than research code

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning, llm-training
- domain: deep-learning, large-language-models, artificial-intelligence, machine-learning
- platform: python
- tags: hierarchical-reasoning-model, recurrent-architecture, reasoning, brain-inspired-ai, arc-benchmark, research-code, pytorch, gpu, linux

## Member repositories
- sapientinc/HRM (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:59.429674+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:13:55.528668+00:00, confidence not recorded.
  - readme: https://github.com/sapientinc/HRM (fetched 2026-08-28T04:10:59.429674+00:00, sha 7ae27f345f59)
  - homepage: https://sapient.inc (fetched 2026-08-29T08:10:22.796971+00:00, sha 16e245dbc9d0)
  - site_page: https://sapient.inc/en/about (fetched 2026-08-29T08:10:22.800001+00:00, sha dbbdb3e5a6b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.