Processing or generating 1,000,000 words (approximately 1,333,000 tokens) with GPT-4 Turbo ranges from $6.665 (cached input) to $39.99 (full generation).
You send 1,000,000 words as prompt context, documentation, or background knowledge.
GPT-4 Turbo drafts a complete 1,000,000 words article, chapter, or code module.
800,000 words input prompt + 200,000 words output reply.
| Model | Provider | Input Cost (1,000,000 words) | Cached Input Cost | Output Cost (1,000,000 words) | Context Limit |
|---|---|---|---|---|---|
| GPT-4 Turbo (Current) | openai | $13.33 | $6.665 | $39.99 | 128,000 |
| GPT-3.5 Turbo | openai | $0.6665 | $0.1666 | $1.999 | 16,385 |
For GPT-4 Turbo (cl100k_base tokenizer), 1,000,000 words is approximately 1,333,000 tokens (an average ratio of 1.33 tokens per word in English). Code, technical vocabulary, and non-English scripts will have higher token densities.
Sending 1,000,000 words as input costs $13.33 (or $6.665 with prompt caching). Generating 1,000,000 words as output costs $39.99. 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 $6.665). For non-urgent asynchronous batch processing, 24-hour batch queues provide a flat 50% discount across all tokens (costing $6.665 for input, $19.995 for output).
At an average human reading speed of 250 words per minute, 1,000,000 words takes approximately 4000 minutes to read. In contrast, GPT-4 Turbo can process or generate this text in seconds.