{"version":"1.0.113","anchor":"per-model-prompt-caching-control","canonical_anchor":"per-model-prompt-caching-control","heading":"Per-Model Prompt Caching Control","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/1.0.113\/e\/per-model-prompt-caching-control","release_url":"https:\/\/changelogs.core-directive.com\/v\/1.0.113","markdown":"### Per-Model Prompt Caching Control\n\n**What:** Fine-grained control over prompt caching for individual model families **How to use:**\n\n`# Disable caching for Haiku models only export DISABLE_PROMPT_CACHING_HAIKU=true # Disable caching for Sonnet models only export DISABLE_PROMPT_CACHING_SONNET=true # Disable caching for Opus models only export DISABLE_PROMPT_CACHING_OPUS=true # Or disable globally for all models (existing behavior) export DISABLE_PROMPT_CACHING=true` **Details:**\n\n- Allows optimization strategies per model tier (e.g., disable for cheaper Haiku, enable for expensive Opus)\n- The global `DISABLE_PROMPT_CACHING` variable still works as before\n- Enables cost and performance tuning based on your specific use cases\n- **Evidence**: `Sd(A)` function at line 395280 replaces `Pd()` at line 394932 in v1.0.112"}