High-performance frontier model for production coding and complex agents
| Scenario | Input Tokens | Output Tokens | Uncached Cost | With Prompt Caching |
|---|---|---|---|---|
| Short Chat Query | 1,000 | 500 | $0.0105 | $0.00770 |
| Document Summarization | 10,000 | 2,000 | $0.0600 | $0.0320 |
| Codebase & Context Analysis | 100,000 | 20,000 | $0.6000 | $0.3200 |
| Batch Corpus Processing | 1,000,000 | 100,000 | $4.50 | $1.70 |
Ideal for production workloads demanding flagship capabilities, deep context depth (200k tokens), and reliability from Anthropic. Excellent when predictable tokenomics and prompt caching support are paramount.
If your use-case requires sub-second streaming latency or ultra-high frequency classification at micro-cent pricing, consider lighter budget options such as Gemini Flash-Lite or Claude Haiku. For deep formal logic, consider dedicated reasoning models like o3.
For Claude Sonnet 5, 1 million input tokens costs $3.00, while 1 million output tokens costs $15.00. If using prompt caching, repetitive input prefixes are discounted to $0.20 per million.
Claude Sonnet 5 features a maximum context window of 200,000 tokens (~150,000 words), with a maximum output limit of 16,384 tokens per completion.
Claude Sonnet 5 utilizes the Anthropic Claude BPE (~65k vocabulary). Token counting on TokenMath runs client-side to ensure maximum privacy.
Yes. Claude Sonnet 5 supports prompt caching with a cached input rate of $0.20/1M (saving up to 93% on repeated input context).