Expired50% confidenceBenchmarkExact time
According to information circulating in the community, GPT-6 achieved a high score on ARC-AGI-3 using a specialized reasoning harness, with the bare model capability reaching approximately 60%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/5/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/19/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedEach GPT-5.6 model can be configured with different thinking intensities, with the highest level being Ultra.71% similarUnverifiedFor the most difficult coding tasks, GPT 5.5 and Opus 4.8 currently each account for approximately 50% of usage, whereas a few months ago Anthropic's models held a clear advantage.69% similarUnverifiedGPT-5.6 Sol achieved a score of 92.5% on ARC-AGI-2.67% similarUnverifiedIn OpenAI's internal benchmarks, GPT-5.6 Sol leads GPT-5.5 by approximately 16 points on recursive self-improvement (model improvement tasks), and shows a 31% improvement over 5.5 on AI infrastructure optimization (chip design workflows).67% similarUnverifiedGPT-5.6 offers native multi-agent support, allowing models to spin up sub-agents for parallel focused work.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/861858API
curl https://kongchang.com/api/v1/knowledge/claims/861858MCP
get_claim(id=861858)