Azure
MindStudio's September 4, 2026 benchmark analysis finds that GPT-6 Astra is not consistently better than Anthropic's Claude Fable 5.1, beating it on computer-use, terminal-workflow, and long-context benchmarks while trailing on broader indices. On Artificial Analysis's Intelligence Index, Fable 5.1 scores 66 versus Ast MindStudio also reports that the headline ARC-AGI-3 score of 99.9% came from OpenAI's provider-adapter harness, which preserves reasoning state between actions, whereas the standard neutral harness put Astra at 62.7%. It lists per-token pricing of $10/$50 (input/output) per million tokens versus $4/$20 for predecessor