Evaluation & BenchmarksFrançois Chollet / ARC PrizeFrançois CholletARC Prize

ARC Prize releases o3's ARC-AGI results

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

o3's ARC-AGI results released — 75.7% in high-efficiency mode ($26 per task), 87.5% in low-efficiency mode ($4,560 per task). At the same time, stated explicitly that o3 is not yet AGI

Why it matters

Created the practice of reporting capability and cost together

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

Officialarcprize.orghttps://arcprize.org/blog/oai-o3-pub-breakthrough

Added in Week 35, 2026 · See that week

On the map — ⑯ Evaluation line

  1. Previous stopGSM1k
  2. This recordo3's ARC-AGI results
  3. Next stopTechCrunch reports criticism of Epoch AI for not disclosing OpenAI funding
See it on the map

Same week

Brian J Chan et al. present Cache-Augmented GenerationPaperDeliberative AlignmentPaperBuilding Effective AI AgentsOfficialGemini 2.0 Flash Thinking ExperimentalPress

Nearby on the same axis

TechCrunch reports criticism of Epoch AI for not disclosing OpenAI fundingPressEpoch AI posts a clarification on how FrontierMath was made and is usedOfficialEpoch AI releases FrontierMathPaperARC Prize announces ARC-AGI-2 and ARC Prize 2025Official