My effective cost per million tokens: Sonnet 5 $0.26, Fable 5.1 $0.69, Opus 5 $0.72. Has anyone measured the same for OpenAI models?

A user shared their effective cost per million tokens for Anthropic models, noting Sonnet 5 at $0.26, Fable 5.1 at $0.69, and Opus 5 at $0.72. The close pricing of Fable and Opus, despite different list prices, is attributed to caching. Anthropic requires explicit opt-in for caching with TTLs and charges 1.25x or 2x input to write entries, while OpenAI caches automatically above 1024 tokens, charging 1.25x to write since GPT 5.6 without reporting writes. The user seeks comparable effective rates for OpenAI models.

This report uniquely details effective LLM costs, showing how caching makes Opus and Fable nearly identical in price despite different list prices, unlike OpenAI's caching approach.