Agents & InferenceSimon Willison

GPT-6.2 Astra costs $8/MTok input and $40/MTok output

Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated.

Match the models (Optional)

Which model wrote which summary? Select a matchup mapping below before voting.

Summary A

GPT-6.2 Astra cuts input/output costs by 20% ($8/$40 per MTok) and cache costs by 60% ($0.40/million), making long-running agentic workflows significantly cheaper. This lets you scale multi-step reasoning or tool-use sessions without proportional cost spikes, but the model may occasionally stall agents with confusing internal-detail questions, requiring tighter prompt guardrails or fallback logic.