All model prices

Claude Sonnet 4.5 (latest)

Vertex AI · claude-sonnet-4-5

Claude Sonnet 4.5 (latest) from Vertex AI: $3 per million input tokens, $15 per million output tokens, $0.3 per million cached input tokens. Retired 29 Sep 2026.

Prices in US dollars per million units
RatePriceApplies
Input$3 per 1M input tokensOrdinary
Cached input$0.3 per 1M cached input tokensOrdinary
Cache write$3.75 per 1M cache write tokensOrdinary
Output$15 per 1M output tokensOrdinary
Input$3.3 per 1M input tokensREGIONAL
Input$1.5 per 1M input tokensBATCH tier
Input$1.65 per 1M input tokensBATCH tier, REGIONAL
Input$6.6 per 1M input tokensprompts over 200,000 tokens, REGIONAL
Input$6 per 1M input tokensprompts over 200,000 tokens
Cached input$0.33 per 1M cached input tokensREGIONAL
Cached input$0.15 per 1M cached input tokensBATCH tier
Cached input$0.165 per 1M cached input tokensBATCH tier, REGIONAL
Cached input$0.66 per 1M cached input tokensprompts over 200,000 tokens, REGIONAL
Cached input$0.6 per 1M cached input tokensprompts over 200,000 tokens
Cache write$6 per 1M cache write tokenscache kept 3,600 seconds or more
Cache write$6.6 per 1M cache write tokenscache kept 3,600 seconds or more, REGIONAL
Cache write$4.125 per 1M cache write tokensREGIONAL
Cache write$1.88 per 1M cache write tokensBATCH tier
Cache write$2.068 per 1M cache write tokensBATCH tier, REGIONAL
Cache write$13.2 per 1M cache write tokensprompts over 200,000 tokens, cache kept 3,600 seconds or more, REGIONAL
Cache write$8.25 per 1M cache write tokensprompts over 200,000 tokens, REGIONAL
Cache write$7.5 per 1M cache write tokensprompts over 200,000 tokens
Cache write$12 per 1M cache write tokensprompts over 200,000 tokens, cache kept 3,600 seconds or more
Output$16.5 per 1M output tokensREGIONAL
Output$8.25 per 1M output tokensBATCH tier, REGIONAL
Output$7.5 per 1M output tokensBATCH tier
Output$24.75 per 1M output tokensprompts over 200,000 tokens, REGIONAL
Output$22.5 per 1M output tokensprompts over 200,000 tokens
Released
29 Sep 2025
Retired
29 Sep 2026
Most input tokens
200,000
Most output tokens
64,000