GPT 6 ASTA obliterated the ARC-AGI-3. when it was launched models struggled to get to 10%. now Astra scores: > 62.7% with the standard harness > 99.9% with provider adapter harness
Hesam(@Hesamation)85 字 (约 1 分钟)
65
GPT 6 ASTA在ARC-AGI-3测试中表现显著提升，标准测试得分62.7%，适配器测试达99.9%。
入选理由：GPT 6 ASTA在ARC-AGI-3标准测试中得分62.7%
FeaturedTweet#GPT#ARC-AGI-3#AI模型#性能测试英文