GPT-6 Astra benchmark results diverge, while ARC-AGI-3 efficiency surpasses humans and prompts Chollet to move up his AGI forecast
Benchmarks disagree on GPT-6 Astra, but its human-beating efficiency on ARC-AGI-3 pulls Chollet’s AGI forecast forward
AI BriefingBenchmark conclusions for GPT-6 Astra conflict: Epoch AI ranked it first among 267 models with a score of 169, while Artificial Analysis gave it 61, on par with its predecessor Sol and behind Claude Fable 5.1 at 66.
Source: The Decoder: AI News · the-decoder.com
Event follow-up1
View all event coverage- GPT-6 Astra benchmark results diverge, while ARC-AGI-3 efficiency surpasses humans and prompts Chollet to move up his AGI forecastThe Decoder: AI News · September 4