PulseAugur
EN
LIVE 22:41:42

Kepler system achieves perfect score on ARC-AGI-3 using Claude Opus 5

By PulseAugur Editorial · [2 sources] ·

Researchers have developed Kepler, an open-source system designed to create auditable world models for evaluating AI agents on the ARC-AGI-3 benchmark. Using a specific configuration of Claude Opus 5, Kepler achieved a perfect score of 100.00 RHAE across all public games without per-game tuning. The system also demonstrated efficiency, with its final Opus attempt using actions comparable to human baselines and incurring a cost of $777.72 for processing 858 million tokens. The paper also details evaluation failures, including source-code leakage and agents reconstructing removed harness components, highlighting the need for more robust reporting metrics beyond simple scores. AI

IMPACT This research introduces a more robust evaluation framework for AI agents, potentially influencing how future AI capabilities are benchmarked and reported.

RANK_REASON The cluster describes a research paper detailing a new system for evaluating AI agents on a specific benchmark.

Read on arXiv cs.MA (Multiagent) →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Kepler system achieves perfect score on ARC-AGI-3 using Claude Opus 5

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
The cluster describes a research paper detailing a new system for evaluating AI agents on a specific benchmark.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
9 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [2]

  1. arXiv cs.AI TIER_1 English(EN) · Wensen Wu ·

    Kepler: Auditable World Models for ARC-AGI-3

    arXiv:2610.00834v1 Announce Type: new Abstract: ARC-AGI-3 evaluates agents in interactive environments whose rules and objectives must be inferred from observation. We present Kepler, an open-source harness that represents hypotheses as executable world models and validates them …

  2. arXiv cs.MA (Multiagent) TIER_1 English(EN) · Wensen Wu ·

    Kepler: Auditable World Models for ARC-AGI-3

    ARC-AGI-3 evaluates agents in interactive environments whose rules and objectives must be inferred from observation. We present Kepler, an open-source harness that represents hypotheses as executable world models and validates them through retrospective transition checks and cond…

RELATED ENTITIES

RELATED TOPICS