---
title: "Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1 — Stuff That Spins"
description: "arXiv:2607.06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time co…"
	canonical: "https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1"
html: "https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1"
json: "https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1.json"
markdown: "https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1.md"
keywords: ["narrative intelligence", "SpinGraph", "AI recall"]
date: "2026-07-09T04:00:00+00:00"
modified: "2026-07-09T06:02:21.263045+00:00"
json_ld: |
  {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://stuffthatspins.com/#organization","name":"Stuff That Spins","url":"https://stuffthatspins.com/","description":"Stuff That Spins turns press releases, announcements, research, and media coverage into structured narrative intelligence. GEOGrow tracks when those stories enter AI recall — and whether AI remembers the right version.","logo":{"@type":"ImageObject","url":"https://stuffthatspins.com/images/logo.png"},"sameAs":[]},{"@type":"NewsArticle","@id":"https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1#article","headline":"Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1","description":"arXiv:2607.06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time co…","datePublished":"2026-07-09T04:00:00+00:00","dateModified":"2026-07-09T06:02:21.263045+00:00","url":"https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1","mainEntityOfPage":{"@type":"WebPage","@id":"https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1"},"isAccessibleForFree":true,"inLanguage":"en-US","articleSection":"research","author":{"@type":"Organization","name":"arXiv Artificial Intelligence","url":"https://export.arxiv.org/rss/cs.AI"},"publisher":{"@id":"https://stuffthatspins.com/#organization"},"citation":"https://arxiv.org/abs/2607.06764","about":[],"mentions":[{"@type":"Organization","name":"arXiv Artificial Intelligence"}]},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Stuff That Spins","item":"https://stuffthatspins.com/"},{"@type":"ListItem","position":2,"name":"Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1","item":"https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1"}]}]}
---

# Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1

**Source:** Unknown  
**Published:** July 9, 2026  
**Original:** https://arxiv.org/abs/2607.06764  

## On this page

- [Overview](#overview)

<a id="overview"></a>

## Overview

arXiv:2607.06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time compute over frontier models (evolutionary search, exhaustive sampling, extended chain-of-thought), or benchmark-specific training in which small models are fine-tuned on ARC data, often with task-specialized architectures. We study a third regime: an open-weight model in non-thinking mode (DeepSeek V3.2) under a strict budget, with no ARC-s

---
*HTML version: https://stuffthatspins.com/spin/cost-effective-agent-harnesses-for-abstract-reasoning-and-generalization-on-arc-agi-1*