Skip to content
@narelabs

Nare Labs

We are a research laboratory dedicated to building the infrastructure for autonomous engineering.

Popular repositories Loading

  1. NARE-CLI NARE-CLI Public

    High-performance CLI agent built for massive refactorings.

    Python 11 1

  2. dsm dsm Public

    Dynamic Segmented Memory (DSM): A hierarchical graph-based retrieval engine for scalable LLM context management and associative reasoning.

    Python 3

  3. sr-moa sr-moa Public

    Self-Reflective Mixture of Adapters (SR-MoA): Autonomous Test-Time Neuroplasticity for Frozen LLMs in PyTorch.

    Python 3

  4. Neuro-Adaptive-Reasoning-Engine Neuro-Adaptive-Reasoning-Engine Public

    NARE CLI Core.

    Python 2

  5. s-git s-git Public

    s-git: Semantic version control that tracks code meaning and AST structure, not text.

    Python 2

  6. ossa ossa Public

    Sparse attention for frozen LLMs. Train a tiny router that picks the top-K keys per query, swap dense attention for sparse — no retraining the model.

    Python 2

Repositories

Showing 10 of 15 repositories
  • mflm Public

    MFLM: A sub-quadratic alternative to Transformers using causal mass-field convolutions for infinite context and O(N) complexity.

    narelabs/mflm's past year of commit activity
    Python 1 MIT 0 0 0 Updated Aug 13, 2026
  • sharrowkin-cli Public

    An autonomous AI coding agent that lives in your terminal.

    narelabs/sharrowkin-cli's past year of commit activity
    TypeScript 1 0 0 0 Updated Jun 2, 2026
  • sharrowkin-installer Public

    Installer for Sharrowkin IDE

    narelabs/sharrowkin-installer's past year of commit activity
    0 MIT 0 0 0 Updated Jun 1, 2026
  • danius Public

    DANIUS-1: Teaching frozen language models to think in shapes and grids, not words. A project by Nare Labs.

    narelabs/danius's past year of commit activity
    Python 2 MIT 0 0 0 Updated May 28, 2026
  • amlb Public

    A decoupled context-compression co-processor utilizing gated cross-attention for linear-time, O(1)-memory sequence scaling in frozen large language models. (159 символов)

    narelabs/amlb's past year of commit activity
    0 0 0 0 Updated May 27, 2026
  • ossa Public

    Sparse attention for frozen LLMs. Train a tiny router that picks the top-K keys per query, swap dense attention for sparse — no retraining the model.

    narelabs/ossa's past year of commit activity
    Python 2 Apache-2.0 0 0 0 Updated May 26, 2026
  • SQT Public

    Static-Q Transformer: replace self-attention's content-based query projection with a learnable per-position table. Empirical study of where this works and where it breaks across model sizes.

    narelabs/SQT's past year of commit activity
    Python 0 0 0 0 Updated May 26, 2026
  • CARG Public

    Capability-Augmented Generation through Reusable Graph — a research framework for accumulating and composing executable LLM skills.

    narelabs/CARG's past year of commit activity
    Python 1 0 0 0 Updated May 25, 2026
  • persistent-cognition Public
    narelabs/persistent-cognition's past year of commit activity
    Python 1 MIT 0 0 0 Updated May 25, 2026
  • s-git Public

    s-git: Semantic version control that tracks code meaning and AST structure, not text.

    narelabs/s-git's past year of commit activity
    Python 2 MIT 0 0 0 Updated May 17, 2026
View all repositories

People

This organization has no public members. You must be a member to see who’s a part of this organization.