Unverified50% confidenceBenchmarkExact time
GPT-6 Astra achieved a near-perfect score of 99.9% on ARC-AGI-type tests, while the average score of ordinary human test-takers was only 48%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/18/2026
First Seen
Valid until: 12/17/2026
Sources
Related Entities
Related Claims
UnverifiedARC-AGI-2 was designed by AI safety researcher François Chollet; early GPT-4 scored less than 10% on this test, while the human average score is approximately 85%76% similarUnverifiedOn the ARC-AGI-3 benchmark, GPT-5.6 Sol scored 8%, while most other models scored below 2%; Anthropic did not run Fable due to prohibitively high costs.72% similarUnverifiedGPT-4 as a judge can achieve over 80% agreement with human evaluations, but exhibits systematic issues such as position bias, verbosity bias, and self-preference.68% similarUnverified2023年人类在GAIA测试中的平均成功率约为90%，而当时最强的GPT-4在Level 1上仅勉强达到15%65% similarUnverifiedGPT-4等大模型在Spider等基准测试上的零样本或少样本SQL生成准确率已超过80%64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/932462API
curl https://kongchang.com/api/v1/knowledge/claims/932462MCP
get_claim(id=932462)