LLM API Pricing Comparison: 5 Traps in Per-Million-Token Prices
How to compare LLM API pricing without a stale table: input vs. output cost, output-to-input ratio, context, caching, batching, tokenizers, plus a checklist.
Blog
#cost optimization
与「cost optimization」相关的文章。
Claude Prompt Caching: Where to Mark, Verify Hits, Check Bills
How Claude prompt caching works, which requests benefit, where to put cache_control, how to confirm hits from the usage fields, plus a pre-launch checklist and a cost script.
DeepSeek API Off-Peak Pricing: Beijing Time Hours and Monthly Cost
DeepSeek API has billed peak and off-peak rates since Aug 16, with off-peak at half price. Get a Beijing time table, a cron example and a monthly cost estimate.