Verified65% confidenceBenchmarkExact time
Under the standard framework of ARC Prize, GPT-6 Astra's ARC-AGI-3 score is only 62.7.
3
Sources
65%
Confidence
Long-term
Relevance
9/21/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the ARC-AGI-3 benchmark, the same GPT-5.6 model with only two configuration items in the Codex Harness changed saw its score rise from 13.3% to 38.3%, with output tokens reduced by a factor of six.71% similarUnverifiedPreviously, top models such as GPT-4 had scores on ARC-AGI that hovered in the range of single digits to 30 for a long time.67% similarUnverified在Terminal Bench 2.1基准测试上，GPT-5.6 Sol的Ultra模式接近92%，超越竞品的88%表现64% similarUnverifiedOpus 5在ARC-AGI 3上的得分是次优模型的三倍64% similarUnverifiedAcross four rounds of testing, Astra won three and tied on dynamic graphics, leading the reviewer to judge GPT-6 as the overall winner.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/938340API
curl https://kongchang.com/api/v1/knowledge/claims/938340MCP
get_claim(id=938340)