Processing or generating 25,000 words (approximately 33,000 tokens) with Claude Sonnet 5 ranges from $0.0066 (cached input) to $0.33 (full generation).
You send 25,000 words as prompt context, documentation, or background knowledge.
Claude Sonnet 5 drafts a complete 25,000 words article, chapter, or code module.
20,000 words input prompt + 5,000 words output reply.
| Model | Provider | Input Cost (25,000 words) | Cached Input Cost | Output Cost (25,000 words) | Context Limit |
|---|---|---|---|---|---|
| Claude Sonnet 5 (Current) | anthropic | $0.066 | $0.0066 | $0.33 | 1,000,000 |
| GPT-5.6 Terra | openai | $0.064 | $0.0064 | $0.384 | 1,050,000 |
| GPT-5.4 Workhorse | openai | $0.08 | $0.008 | $0.48 | 256,000 |
| o3-mini | openai | $0.0352 | $0.0176 | $0.1408 | 200,000 |
| o4-mini | openai | $0.0352 | $0.0088 | $0.1408 | 256,000 |
| o1-mini | openai | $0.0352 | $0.0176 | $0.1408 | 128,000 |
For Claude Sonnet 5 (claude_bpe tokenizer), 25,000 words is approximately 33,000 tokens (an average ratio of 1.32 tokens per word in English). Code, technical vocabulary, and non-English scripts will have higher token densities.
Sending 25,000 words as input costs $0.066 (or $0.0066 with prompt caching). Generating 25,000 words as output costs $0.33. Output tokens are more expensive because autoregressive token generation requires significantly more computation per token.
Yes! If your input text is part of a repeated context, prompt caching saves 50% to 90% (costing $0.0066). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $0.033 for input, $0.165 for output).
At an average human reading speed of 250 words per minute, 25,000 words takes approximately 100 minutes to read. In contrast, Claude Sonnet 5 can process or generate this text in seconds.