Public KG record
prompt cache
Recorded overview
reused after a fork to avoid re-paying prefix steps.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Capabilities: cache prewarming
- Capabilities: explicit cache breakpoints
- Capabilities: Prompt caching diagnostics
- Performance: 90-99% hit rates
- Performance: 新鲜处理的 prompt token 占比减少超过50%
- Performance: 缓存输入token成本最高比未缓存低95%
- Pricing: 缓存输入 token 最高可享 90% 折扣
- Pricing: 缓存输入token最高可享90%折扣
Evidence coverage
The KG currently records 9 known information dimensions and has 10 undocumented dimensions, linked to 6 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- GitHub Copilot — uses
- Manus — uses
- OpenAI — develops
- Agent Harness — has feature
- API — supports