Laya: Fine-Tuning an Open-Weight Decision Model for Skill Routing

VideoCURTDemo

CURT explores Laya, an open-weight Jev-style decision model, using it to shortlist skills for Hermes Agent rather than relying on frontier models for every routing decision.
Playing loads YouTube (youtube-nocookie.com).

Key points

  1. Laya is an open-weight decision model structured around typed decisions and option scoring.
  2. The experiment benchmarks Laya against Jev on shortlisting skill candidates for Hermes Agent sessions.
  3. Laya is fine-tuned on historical routing logs and tested against held-out messages.
  4. The author conducts an expanded training run to evaluate changes in ranking quality with more data.
  5. The model acts strictly as a candidate-filtering layer rather than a direct replacement for the agent's ultimate skill choice.

Why it matters

Filtering tools with lightweight, specialised open-weight models helps lower latency and API overhead in multi-skill agent workflows.

Analysis

This setup highlights the benefit of decoupling tool pre-selection from execution, using a compact decision model to constrain context before the main agent acts.

Mentioned links

Links and tools from this entry

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Repository:NandhaKishorM/laya

    Laya: Non-Autoregressive Decision Engine for Fast In-Text Choices

    Laya is an open-source Python decision engine designed to deliver typed decisions—such as categorical choices, scores, or yes/no answers—across more than 100 languages. Instead of relying on autoregressive text generation, it produces decisions in a single forward pass taking approximately 33 ms.

    26kstarsPython

    AI & LLMs· Tool

  • Video

    Video:CURT

    Building a Local Jev-Inspired Classifier with Claude Opus 5.5

    CURT demonstrates how to use Claude Opus 5.5 to fine-tune a local specialist classification model inspired by Jev. The system categorises travel and hotel features against user requirements and integrates image recognition to verify classifications.

    AI & LLMs· Tutorial

  • Repository:NousResearch/hermes-agent

    Hermes Agent: Self-Improving Open Source AI Agent by Nous Research

    Hermes Agent is a cross-platform, self-improving AI agent designed with an autonomous learning loop, cross-session memory, and integration into multiple messaging platforms and terminal interfaces.

    240.8kstarsPython

    AI & LLMs· Tool

  • Link:Nous Research

    Hermes Agent: Self-Improving Autonomous AI Agent by Nous Research

    Nous Research has documented Hermes Agent, an autonomous AI agent featuring a built-in learning loop. It develops and refines skills through operational experience, retains context across sessions, and operates across diverse environments from local machines to serverless setups.

    AI & LLMs· Resource

  • Video

    Video:CURT

    Jev Design Patterns: Operating Workflows with Decision Models

    This video introduces Jev by TypeSafe, a specialised decision model designed for low-cost, low-latency micro-decisions rather than text generation. It covers three decision types, ten functional shapes, and four architectural design patterns through practical examples.

    AI & LLMs· Tutorial

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.