Tags

caching

Prompt caching and response caching

Topics

• Cost and cachingLLM costs are driven by tokens: input, output (including thinking), and cached input. Most savings come from not reprocessing the same context, choosing the right effort and model for each task,…

Sources

• Gemini 4 Argon model page (Gemini API docs)docs · Google · 13 October 2026
• Claude Code releases 2.1.299 and 2.1.300release · Anthropic · 14 October 2026
• Claude API pricingdocs · Anthropic · 7 October 2026
• Gemini Developer API pricingdocs · Google · 7 October 2026
• OpenAI API models and pricing: GPT-6 familydocs · OpenAI · 2026-10
• DeepSeek API: models and pricingdocs · DeepSeek · 10 September 2026
• Prompt caching (Claude API docs)docs · Anthropic