{"slug": "openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom", "title": "OpenAI claims GPT-5.6 Sol beats Opus 5 on ARC-AGI-3 but only with its own custom test harness", "summary": "OpenAI claims its GPT-5.6 Sol model scored 38.3 percent on the ARC-AGI-3 benchmark, surpassing Anthropic's Opus 5 at 30.2 percent, but only when using OpenAI's own API with retained reasoning and context compaction. In the official test environment, GPT-5.6 Sol managed just 7.8 percent, while Opus 5 achieved its score without such aids.", "body_md": "OpenAI counters Anthropic's ARC-AGI-3 record: GPT-5.6 Sol scores 38.3 percent, but only through its own API with retained reasoning and context compaction. In the official test environment, the model managed just 7.8 percent. Opus 5 hit its 30.2 percent without such aids.\n\nThe article [OpenAI claims GPT-5.6 Sol beats Opus 5 on ARC-AGI-3 but only with its own custom test harness](https://the-decoder.com/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom-test-harness/) appeared first on [The Decoder](https://the-decoder.com).", "url": "https://wpnews.pro/news/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom", "canonical_source": "https://the-decoder.com/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom-test-harness/", "published_at": "2026-07-30 07:40:11+00:00", "updated_at": "2026-07-30 08:08:23.766431+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research", "ai-products"], "entities": ["OpenAI", "GPT-5.6 Sol", "Anthropic", "Opus 5", "ARC-AGI-3"], "alternates": {"html": "https://wpnews.pro/news/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom", "markdown": "https://wpnews.pro/news/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom.md", "text": "https://wpnews.pro/news/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom.txt", "jsonld": "https://wpnews.pro/news/openai-claims-gpt-5-6-sol-beats-opus-5-on-arc-agi-3-but-only-with-its-own-custom.jsonld"}}