Unverified50% confidenceFactExact time
Anthropic在Claude 3系列中将上下文窗口扩展至200K Token并研发了提示词缓存技术，可将实际账单成本降低50%以上
1
Sources
50%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
Claude Sonnet 5接入Devin：编程性能升级，配额消耗降低30%
rss6/30/2026
Related Claims
UnverifiedAnthropic introduced a native prompt caching feature in the Claude series of models that caches and reuses repeated prefix tokens, and a cache hit can save up to 90% of the cost for the corresponding tokens.67% similarUnverifiedThe company claims that the TokenFold architecture can reduce token consumption by 50%, saving 40% in usage costs compared to Claude Code, Codex, or Cursor.62% similarUnverifiedAnthropic's Claude API offers a Prompt Caching feature that reduces the input price by 90% on cache hits.61% similarUnverifiedClaude's input tokens are typically approximately 3–5 times cheaper than output tokens, because the generation process requires autoregressive token-by-token inference.60% similarUnverifiedClaude Code combined with Anthropic's Prompt Caching mechanism can save approximately 80% of Token costs.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/212302API
curl https://kongchang.com/api/v1/knowledge/claims/212302MCP
get_claim(id=212302)