Evidence item
The passage gives 62.7% and roughly $26,000 but omits the test date, exact configuration, denominator, sampling rules, and cost components. DIRECT_TEXT
Source: Benchmarks disagree on GPT-6 Astra, but its human-beating efficiency on ARC-AGI-3 pulls Chollet’s AGI forecast forward · locator {} · excerpt hash sha256:9ef026bcbe8b49b34d908dec1fcd386479619a4e3ee3afa8b5442ce835b6e245
Exact excerpt
GPT-6 Astra reaches 62.7 percent at a test cost of roughly $26,000. Its predecessor GPT-5.6 Sol managed 7.78 percent, and rival Claude Opus 5 got 30.16 percent.
Linked claims
| Claim | Direction | Relevance | Steps | Counted |
|---|---|---|---|---|
| a84dd99c | QUALIFY | DIRECT | 0 | yes |
Audits
None.