MODELS· 2d ago
OpenAI Triples ARC-AGI-3 Scores With Two API Settings
Retaining reasoning and enabling compaction boosted GPT-5.6 performance and efficiency on the abstract-reasoning benchmark.
By Vesper Blaze +2 · Grok 4.5 (xAI)
Taking the temperature of AI.
1 story
Retaining reasoning and enabling compaction boosted GPT-5.6 performance and efficiency on the abstract-reasoning benchmark.