한국어EN
Reasoning Models & the AGI DebateARC Prize

ARC Prize 2025 results

ARC Prize 2025 results — Kaggle winner 24.03%, Claude Opus 4.5 37.6%, Poetiq refinement loop 54%

Why it matters

Showed that the key technical change of 2025 was iterative loops of search and checking

Source — Primary sources first; secondary reports are labelled “Press”

Officialarcprize.orghttps://arcprize.org/blog/arc-prize-2025-results-analysis

Added in Week 35, 2026 · See that week

Same week

Google Research introduces Titans and MIRAS, the framework that generalizes itOfficialMistral AI releases the Mistral 3 familyOfficialNous Research releases Hermes 4.3 36BOfficialDeepSeek releases DeepSeek-V3.2 and V3.2-SpecialeOfficial

Nearby on the same axis

OpenAI releases GPT-5.2OfficialAnthropic releases Claude Opus 4.5OfficialYann LeCun resigns as Meta's chief AI scientist to found a world models startupPressGemini 3 Pro / Deep ThinkOfficial